# LLM Full Content
Site: Smart Connections
Created at: 2026-09-10T21:27:26.130Z
## 1 2
canonical: https://smartconnections.app/smart-graph/releases/1-2/
html_url: https://smartconnections.app/smart-graph/releases/1-2/
markdown_url: https://smartconnections.app/smart-graph/releases/1-2.md
llms_url: https://smartconnections.app/smart-graph/releases/1-2/llms.txt
last_modified: 2026-09-10T21:26:54.012Z
usage_notes: |-
Use this page to answer questions about 1 2.
excerpt: |-
Smart Graph Pro v1.2 Turn a cluster into a working set A graph that ends as a picture is a dead end. Smart Graph Pro v1.2 lets you recognize a meaningful region, review the actual notes inside it, and turn that decision into Context, Canvas, a note, links, or a PNG without selecting the sources again. The graph becomes a place to decide what belongs together, not just a different way to look at…
suggested_links:
- title: 1 0
url: https://smartconnections.app/smart-graph/releases/1-0/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
# Smart Graph Pro v1.2
## Turn a cluster into a working set
A graph that ends as a picture is a dead end. Smart Graph Pro v1.2 lets you recognize a meaningful region, review the actual notes inside it, and turn that decision into Context, Canvas, a note, links, or a PNG without selecting the sources again.
The graph becomes a place to decide what belongs together, not just a different way to look at the vault.
> Update all installed Smart Plugins together, then restart Obsidian. Smart Graph Pro v1.2 requires Smart Environment v3.
## A better semantic foundation for the graph
The shape of Smart Graph begins with embeddings, so Smart Environment v3 is a substantial part of this release. Graph gets a faster shared startup path, the Pro accelerated vector index, a broader built-in local model catalog, and the ability to switch embedding models without deleting the embedding set you may want to revisit.
That makes it easier to choose a lighter or multilingual model for a particular vault, compare how the semantic landscape changes, and return to the earlier model without treating the experiment as irreversible.
Shared drag-and-drop and menu actions also make Graph a more natural continuation from Connections, Lookup, Context, or the note already in front of you.
Learn more about the release of [Smart Environment v3](https://smartconnections.app/smart-environment/releases/3-0/).
## 1. Find the region that matters
Select one note, several nodes, or a semantic cluster. Search while non-matches dim, drop indexed notes or folders onto the graph, or send a Smart Lookup result set into a new visual scope.
The surrounding shape remains visible, so bridge notes, isolated ideas, and neighboring clusters can change your understanding without changing the exact working set you are building.
## 2. Review the sources as notes
The new Notes panel turns the visual selection into a readable list. Selected clusters and notes remain grouped, and you can open a source, remove one item, clear the selection, or move the panel out of the way. Minimized and collapsed states keep the graph usable in smaller panes.
Link previews, authored-link display, graph reveal, and semantic-cascade replay help explain the region without forcing you to leave the visual and reconstruct the list elsewhere.
## 3. Turn the decision into work
From the same reviewed selection or scope, you can:
- Add the sources to Smart Context.
- Create an Obsidian Canvas.
- Create a source-linked note, with optional synthesis support.
- Copy selected or scoped links.
- Save the graph as a PNG.
One source set can become a spatial workspace, a written draft, reusable Context, or a shareable image without being selected again.
## Before / After
| Before | With Smart Graph Pro v1.2 |
| --- | --- |
| A useful visual cluster often ended as a screenshot or memory. | Preserve the exact selection as Context, Canvas, note, links, or PNG. |
| It was difficult to review a dense region as ordinary notes. | The Notes panel makes the source set readable and editable. |
| Lookup and Graph required a manual handoff. | Send a reviewed Lookup result set directly into a visual scope. |
| Trying another embedding model risked discarding the earlier semantic index. | Smart Environment v3 keeps previous model embeddings available when you switch. |
| Refreshes and pane changes could leave stale graph state behind. | Deterministic refresh and lifecycle cleanup rebuild the graph more reliably. |
## Supporting improvements
- Drag-and-drop visual feedback for indexed notes and folders.
- Search dimming, label prioritization, empty states, and performance reporting.
- More consistent Context, scope, link, and Create menus.
- Deterministic refresh and cleanup across close, reopen, pane movement, and new windows.
- Added the Notes panel for selected clusters and notes, including drag, minimize, collapse, clear, and remove controls.
- Added Canvas, note, optional synthesis, PNG, Context, scope, and link actions for reviewed selections.
- Added drag and drop for indexed items and folders with visual feedback.
- Added Smart Lookup handoff into Smart Graph.
- Added source-link hover, authored-link display, graph reveal, semantic cascade, and replay actions.
- Improved labels, search dimming, empty states, performance reporting, and action-overlay behavior.
- Fixed deterministic refresh and lifecycle cleanup across refresh, close, reopen, pane movement, and new windows.
- Updated Smart Graph Pro to Smart Environment Pro v3.2.0.
- Requires Smart Environment v3.0.0 or newer.
## Learn more
- [Smart Graph overview](https://smartconnections.app/smart-graph/)
- [Smart Graph documentation](https://smartconnections.app/docs/graph/)
- [Smart Graph FAQ](https://smartconnections.app/smart-graph/faq/)
## Release notes
### `v1.2.2`
#### Shape the graph around what matters
Start a graph from a Bases file, dim nodes by how recently their sources were edited, and remove selected nodes from the current graph without deleting the underlying notes. Opening and navigating graphs is also more reliable as you move between selections and panels.
#### Turn a graph selection into a note or Canvas
Move from exploration to a reusable artifact without leaving the graph. Creating a note or Canvas from the current selection has been streamlined.
*Turn the current graph selection into a note or Canvas from the Create menu.*
#### Full release notes
##### Exploration
- Create graph scopes from Bases files.
- Dim nodes according to how recently their sources were edited.
- Remove selected nodes from the current graph scope without deleting the source notes.
##### Interaction and output
- Improved graph-open requests, deferred rendering, and view-state handling.
- Improved Escape-key behavior for clearing selections and managing panels.
- Streamlined Canvas and note creation.
- Updated the Connections-to-graph handoff to remain compatible with the revised Connections integration.
##### Reliability and maintenance
- Added cluster-graph decoding and simplified the related processing.
- Added an explicit error when required seeded-cluster graph support is unavailable in the index, and simplified the related ranking logic.
- Added automated checks for graph replay and PNG capture.
### `v1.2.1`
Updated: Smart Environment
### `v1.2.0`
Improved: compatibility with new windows
Improved: Connections list Graph integration using Smart Environment menu actions pattern
Refactor smart graph label handling and rendering logic
- Moved label-related constants to `smart_graph_utils.js` for better organization.
- Consolidated label measurement and collision detection functions.
- Enhanced label rendering logic to prioritize selected and search match nodes.
- Introduced new utility functions for generating action payloads and render signatures.
- Updated tests to cover new label admission logic and rendering status handling.
- Removed deprecated functions and streamlined label budget calculations.
Refactor smart graph utility functions and add tests for capture filename and search query normalization
Enhance smart graph rendering: add empty state handling and performance logging functions
Refactor smart graph actions: remove legacy code, introduce overlay component, and enhance utility functions
- Deleted `smart_graph_actions.js` and moved relevant functions to `smart_graph_utils.js`.
- Added new `smart_graph_actions_overlay.js` to handle UI rendering and action execution.
- Simplified utility functions for selecting nodes and checking visible actions.
- Updated tests to reflect changes in utility functions and removed obsolete tests.
Add scope management actions: implement graph_add_scope_to_smart_context and graph_copy_scope_links, enhance smart_graph with scope item handling
Add PNG saving functionality: implement graph_save_png action and related utilities for capturing and saving graph images
Refactor render event handling: update payload structure and enhance error reporting in emit_rendered_event
Add notes panel component: implement smart_graph_notes_panel for enhanced graph interaction (displays selected clusters/nodes)
Enhance notes panel state management: implement cluster grouping logic and add unit tests for selection behavior
feat: enhance smart graph notes panel with drag-and-drop functionality and UI improvements
- Added dragging capability to the notes panel for better user interaction.
- Introduced minimized and collapsed states for the notes panel, allowing users to manage space effectively.
- Updated styles for the notes panel header and controls to improve usability and visual feedback.
- Implemented new control buttons for clearing selection and toggling panel states.
- Enhanced accessibility with ARIA attributes for interactive elements.
- Refactored panel rendering logic to accommodate new UI states and improve performance.
feat: add create canvas and create note functionalities with respective menus and actions
- Implemented graph_create_canvas action to create a canvas from selected items.
- Implemented graph_create_note action to create a note summarizing selected items.
- Added menus for creating canvas and notes in the graph actions overlay.
- Enhanced smart_graph_actions_overlay to handle new menu actions.
- Updated smart_env.config.js to include new actions for canvas and note creation.
feat: enhance smart graph notes panel with remove selection functionality
- Added a remove button for selected nodes in the notes panel.
- Implemented logic to handle the removal of selections from the graph.
- Updated styles for the new remove button to ensure consistency with existing controls.
- Enhanced search functionality to dim non-matching nodes and edges during searches.
- Introduced utility functions for managing source selection and search dimming states.
- Added tests to verify the new functionality and ensure existing features remain intact.
feat: implement note creation features with synthesis support and bridge review candidates
feat: enhance context menus with new actions for scope management and copying links
feat: hide save button when it's the only action in the smart graph actions
added: graph animation (reveal)
feat: add hover link functionality for source nodes and style adjustments for source links
Added: show links action
Added: graph reveal animation
Added animation action
Fixed: deterministic Smart Graph refresh
- Forced refresh now disposes the active Cytoscape runtime and replaces its graph container.
- Render revisions and latest-request coalescing prevent superseded work from committing graph, DOM, status, or rendered-event state.
- View close, reopen, and window migration invalidate pending renders and clean up listeners, observers, schedulers, replay state, and search timers.
Improved: Smart Context creation actions
- Moved selection and scope context creation into the Create menu while preserving the existing action keys.
- Hid the scope action when selection and scope resolve to the same source set.
Documented: Show links, Replay graph reveal, and Replay semantic cascade as release-supported actions
Implement graph_add_items_to_scope action and associated tests; enhance smart_graph with drag-and-drop functionality for indexed items
Add lookup_list_send_to_smart_graph action and associated menu for exploring lookup results in Smart Graph
Refactor get_dropped_folder_sources to improve path normalization and matching logic
Add drag-and-drop visual feedback for the smart graph main area
Updated: Smart Environment v3
---
## 3 0
canonical: https://smartconnections.app/smart-environment/releases/3-0/
html_url: https://smartconnections.app/smart-environment/releases/3-0/
markdown_url: https://smartconnections.app/smart-environment/releases/3-0.md
llms_url: https://smartconnections.app/smart-environment/releases/3-0/llms.txt
last_modified: 2026-09-10T21:06:24.855Z
usage_notes: |-
Use this page to answer questions about 3 0.
excerpt: |-
Smart Environment v3 Open faster. Change models without starting over. Smart Environment v3 is the upgrade you feel before opening any individual Smart Plugin. The suite reaches a useful state sooner, offers more built-in local embedding models, and lets you change the active model without restarting Obsidian or erasing the embeddings created by the model you were using before. It also makes the…
suggested_links:
- title: 2 4
url: https://smartconnections.app/smart-environment/releases/2-4/
- title: 2 5
url: https://smartconnections.app/smart-environment/releases/2-5/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
# Smart Environment v3
## Open faster. Change models without starting over.
Smart Environment v3 is the upgrade you feel before opening any individual Smart Plugin. The suite reaches a useful state sooner, offers more built-in local embedding models, and lets you change the active model without restarting Obsidian or erasing the embeddings created by the model you were using before.
It also makes the invisible parts of the system much easier to trust. Notes and source sets move more cleanly between plugins, useful actions appear in more consistent menus, and you can finally inspect what was embedded instead of guessing from the results.
> Smart Environment v3 is a coordinated suite upgrade. Update every installed Smart Plugin, then restart Obsidian. Mixed v2 and v3 environments are blocked so an incomplete update cannot leave the suite in a misleading half-updated state.
## Faster startup across the suite
v3 removes more blocking, repeated, and unnecessary work from the path between opening Obsidian and using a Smart Plugin. Built-in local embedding work can initialize away from the main interface, unchanged sources no longer need the same full traversal on every load, and a local model can remain unloaded when there is no embedding work to perform.
Plugins also share more of the same readiness path instead of repeating setup on their own. The practical result is less time waiting for Connections, Lookup, Context, or Graph to become useful. When something is still loading, the Status view now makes that state clearer instead of leaving an empty surface to explain itself.
### Measured startup and embedding impact
A Smart Connections benchmark exercised the Environment pipeline on one Windows system with 281 sources, 7,341 blocks, and the TaylorAI/bge-micro-v2 local model. It compared Smart Connections 4.5.3 with Smart Environment 2.4.6 against Smart Connections 4.7.2 with Smart Environment 3.1.1.
| Measurement | v2 baseline | v3 observed | Difference |
| --- | ---: | ---: | ---: |
| First load to plugin-ready | 37.68 s | 24.96 s | 33.7% faster |
| Subsequent load to plugin-ready | 5.03 s | 3.35 s | 33.5% faster |
| Local embedding character throughput | 39,830 chars/s | 48,801 chars/s | 22.5% higher |
| Subsequent `smart_sources` load | 1.786 s | 0.138 s | 92.3% shorter |
| Empty embedding-queue path | 1.002 s | less than 1 ms | More than 99.9% shorter |
| Vector preparation, storage, checkpoints, and final commit | 1.357 s | 0.090 s | 93.4% shorter |
On the subsequent load, the older path traversed an import queue containing all 281 sources, initialized the local model, built an embedding queue, and only then discovered that no items required embedding. The v3 path processed the one source requiring import work and returned from the empty embedding queue without creating the model.
During first-load embedding, the older path issued one model call per item. The v3 path uses worker-backed batching, a batch window, and input-length sorting so each model call can process more useful work. Vector-file persistence also avoids much of the repeated source, block, and checkpoint writing performed by the older inline-storage path.
> These measurements are representative observations, not universal guarantees. Only one run per version and condition was captured, and cache state, hardware, model choice, and vault contents can change the result. The v3 first-load queue also contained 19.2% fewer input characters, so its raw 36.2% shorter embedding duration is not an equal-work comparison. The 22.5% character-throughput improvement is the more useful normalized result.
## Switch embedding models without throwing the old index away
Changing embedding models used to feel like a one-way migration. In v3, each model can keep its own stored embeddings. You can activate another model without reloading Obsidian or deleting the data produced by the previous model, then switch back later without treating the experiment as a full reset.
The built-in local model catalog is broader too, including more lightweight and multilingual choices for different vaults and hardware. Embedding and chat model settings now make the active model, readiness, dimensions, errors, deletion, and reindexing easier to understand.
## The suite now behaves like a suite
Drag a saved Context into Chat. Send Lookup results to Graph. Drop a vault note onto Connections. Open the menu on a source and continue from that source without first navigating to another plugin.
Supported notes, blocks, folders, saved contexts, and result sets now arrive as the thing you moved instead of collapsing into a file path or plain text. The same shared action system also gives menus, commands, ribbon buttons, source actions, and status controls more consistent behavior across the suite.
## Stop guessing about the index
The improved Stats view turns index health into something you can inspect. Review source and block totals, embedding coverage, loaded-vector memory, skipped items, and unexpected vectors from one place.
The source inspector goes one level deeper. Open the active note or another source to see what was imported, what was embedded, what was skipped, and why. Search and reason filters help with larger collections, and **Force re-import** gives you a direct recovery path when one note is stale.
## Export what the Environment has built
Data export now lets you choose the collections to include, decide whether vectors belong in the export, watch progress, and see a clear completion or retry state. Backups, diagnostics, and migrations no longer require blind trust in an opaque export step.
## Pro: API keys belong in Obsidian's keychain
Smart Environment Pro can store supported API keys through Obsidian's native, keychain-backed secret storage instead of ordinary plugin settings. Pro plugins get one safer shared way to retrieve credentials while Obsidian remains responsible for protecting the secret.
## The v3 difference
| Before | With Smart Environment v3 |
| --- | --- |
| Smart Plugins could repeat blocking work, revisit unchanged sources, or initialize a local model before discovering that the embedding queue was empty. | Shared startup work is reduced, source processing is incremental, and local models stay unloaded until embedding work exists. |
| Local embedding commonly issued one model call per item and repeatedly persisted inline vectors during longer runs. | Worker-backed, length-aware batches process more items per call while vector-file persistence reduces repeated checkpoint and collection writes. |
| Changing embedding models felt like committing to a rebuild. | Switch models without restarting Obsidian or deleting another model's stored embeddings. |
| The built-in local model choice was narrower. | Choose from a broader catalog, including more lightweight and multilingual options. |
| Dragging between plugins could reduce a useful object to a path or text. | Supported items retain enough meaning for the destination to continue the workflow. |
| Index problems were difficult to locate. | Stats and source inspection show what is embedded, skipped, stale, or unexpected. |
| Exporting shared data required more trust than feedback. | Choose the contents, watch progress, and see a clear completion or retry state. |
## Supporting improvements
- Deferred local embedding-model initialization until an embedding queue contains work.
- Incremental source import on subsequent loads instead of revisiting every unchanged source.
- Worker-backed, length-aware batched local embedding where supported.
- Vector-file persistence and lighter checkpoint commits, reducing repeated source and block writes.
- Removed path-length failures in source and embedding storage, so every note name Obsidian supports now works, even in deeply nested vaults.
- Safer source and embedding saves during unload, with more durable sharded persistence.
- Source-data optimization with backup validation and clearer recovery behavior.
- Batched vector maintenance after source re-imports instead of unnecessary one-at-a-time rebuilds.
- Gitignore-style file and folder exclusions.
- Less disruptive notices with expandable details, help links, and mute controls.
- Better Bases, Canvas, rendered-source, binary-file, image, and PDF handling where supported.
- More accurate Core and Pro vector memory reporting.
## Learn more
- [Smart Plugins documentation](https://smartconnections.app/docs/plugins/)
- [Smart Plugins getting started](https://smartconnections.app/getting-started/)
- [Smart Environment FAQ](https://smartconnections.app/smart-environment/faq/)
## Release notes
### Smart Environment v3.0.0
- Established v3 as the compatibility boundary for the coordinated Smart Plugin release train.
- Prevented Smart Environment 2.x plugins from contributing configuration to, or remaining loaded in, a v3 environment.
- Improved Plugin Store compatibility signaling and documented the update-all-plugins restart path.
### Smart Environment Core
#### v3.1.0
- Reduced startup friction with incremental source processing, worker-backed local embedding initialization, and an empty-queue fast path that avoids loading the model when no vectors are missing.
- Added worker-backed local embedding batches and input-length-aware queue processing, allowing each model call to process more useful work.
- Added non-destructive embedding-model switching, active-model reindexing, better Ollama handling, and clearer model controls.
- Expanded the built-in local embedding catalog with additional lightweight, multilingual, and experimental models.
- Added shared menu, command, ribbon, source, parameter, and drag-and-drop actions across supported plugin surfaces.
- Expanded Stats and source inspection with memory reporting, embedding health, skipped and unexpected item filters, active-note inspection, force re-import, and clearer errors.
- Improved data export and source-data optimization.
- Removed path-length failures in source and embedding storage, so every note name Obsidian supports now works, even in deeply nested vaults.
- Improved sharded persistence, save and unload behavior, queued re-import handling, and subsequent-load source reuse.
- Added Gitignore-style exclusions and improved notifications, plugin installation, release links, Canvas parsing, binary reads, and context compilation.
#### v3.1.3
##### Switch embedding models without starting over
Change embedding models without deleting data saved for models you have already used. You also get more built-in model choices and faster processing for the default local model.
*Compare the expanded built-in model choices before choosing what should index your vault.*
##### See what was skipped - and fix it
Environment stats makes skipped and unexpected items easier to investigate. Search the list, filter by reason, inspect the source behind a result, force a re-import, repair block embeddings, or optimize stored source data with backup validation.
Long source paths no longer get excluded during normal indexing simply because they exceed 200 characters.
*Filter skipped blocks by reason, then inspect the source behind each result.*
##### Keep credentials in secure storage
API keys and legacy OAuth tokens now use shared secure storage, backed by Obsidian's native secret storage where available. Migration checks that secure storage can persist a secret before moving existing credentials.
##### A calmer event feed
Open **View more** when you need the details behind an event, then load incoming events with **Show more** when you are ready. New activity no longer shifts the feed while you are reading.
##### Full release notes
###### Models and indexing
- Switch embedding models without deleting previously saved embedding data.
- The default built-in embedding model now uses background-worker processing for improved performance.
- Improved local embedding batch-size handling and added batch-window and sorting configuration.
- Added more Transformers embedding models and support for selecting model revisions.
- Improved the model settings layout and added confirmation before deleting a model configuration.
- Added a way to re-index embeddings for the active model.
- Deselected blocks are excluded from embedding preparation, with automated checks for block selection.
- Improved embedding saves, memory-capacity reservation, and error handling during embedding processing.
- Pending source and block embeddings are saved before their in-memory stores are cleared when Smart Environment unloads. This lifecycle is now handled in Core.
###### Sources and stored data
- Removed the old 200-character source-path exclusion during normal use, with long paths supported through shared, sharded storage. The V2 per-source filename limit remains only in the one-time legacy migration.
- Improved file-link parsing in Canvas files.
- Added .gitignore exclusions and improved folder-exclusion handling.
- Standardized file and folder exclusion settings as lists and normalized how they are handled.
- Added **Re-import wait time** in Smart Environment settings to control the delay before automatic re-import.
- Added configurable delays for queued collection and event-log saves, with less frequent event-log writes.
- Improved Environment data export.
###### Diagnostics and recovery
- Improved Environment statistics and source inspection, including memory usage for embedding vectors and vector-file storage metrics.
- Added an inspector for skipped and unexpected source and block items, with search and reason filters.
- Collection cards now open item inspection. Inspector buttons, inputs, and accessibility labels have been improved, along with loading states and error handling.
- Source Inspector now offers a force re-import option.
- Environment stats now includes block-embedding integrity checks and repair controls.
- Added source-data optimization to Environment stats, including backup validation and error handling.
- Embedding-error events now include the provider's API response JSON for troubleshooting.
###### Credentials
- API keys and other secrets now use shared secure storage in Core, with Obsidian-native secret storage where available. This support is no longer Pro-only.
- Improved secret migration and legacy OAuth-token handling. Migration now checks that the secure-storage adapter can persist secrets before proceeding.
###### Notifications and navigation
- Notifications now offer **View more** to open event details.
- The events feed uses **Show more** to load new events instead of shifting the content automatically.
- Added notification help links and control over whether a notification shows a mute button.
- Invalid event values no longer crash event dialogs or their displays.
- Improved badge icons and added accessible tooltip labels.
- Plugin installation uses fewer notifications, links to release pages, and fewer unnecessary installation-state checks.
- Improved the Smart Plugins list dialog and header styling in fuzzy-search dialogs.
- Updated the Environment status-bar menu, added an Environment status option, and added buttons to the status view.
- Added a shared command for opening release notes, replacing separate per-view command registration.
- Fixed missing view icons by registering views before the workspace renders.
- Fixed hover navigation between adjacent submenus.
- Updated Obsidian-link handling to the newer protocol API.
###### Shared infrastructure and maintenance
- Added support for reading files as binary bytes.
- Added sharded source storage with numeric replay order, bounded append-file rotation, explicit compaction, and protection for legacy base-last commits. Rotation and compaction limits now use byte sizes instead of record counts.
- Core now handles collection-level binary vector loading and typed-array similarity calculations. Durable vector references (`file_i`) are saved only after the vector data they reference.
- Corrected the scope of vector-index operations and added shared retrieval of the strongest cosine-similarity matches across all embedded items in a collection.
- Standardized embedding-input preparation through shared actions, with dedicated handling for Bases, Canvas, and rendered sources.
- Updated Transformers to version 4.2.0 and removed unused model dependencies.
- Standardized menu registration, resolution, and building, and added shared registration for Command Palette and ribbon actions. Added automated checks for menu behavior.
- Improved parameter and event handling for menus, ribbon actions, Smart Plugins commands, and Environment Status View commands, with automated checks.
- Notification buttons can now use `btn_event_key` and `btn_event_payload` instead of `btn_callback`.
- Improved configuration-version handling and added automated checks for Environment creation.
- Removed an unused collection-settings component, updated shared configuration references, and simplified the internal organization of stats inspection.
- Added automated checks for exclusion rules, long source paths, and legacy filename exclusions.
- Updated to Smart Environment v3 and refreshed related version metadata.
### Smart Environment Pro
#### v3.2.0
- Moved the accelerated vector index and its v3 embedding lifecycle into Smart Environment Pro.
- Added model-scoped vector-file persistence, batched length-sorted embedding, and lighter checkpoint commits that reduce repeated source and block writes.
- Added batched index rebuilds, custom embedding-dimension support, accurate OpenAI dimensions, and combined Core and Pro memory reporting.
- Coordinated with Core's incremental source processing so an empty embedding queue can return without initializing the local model.
- Added Obsidian-native secure secret storage for supported API keys.
- Added per-view Bases rendering, relative `this.file` and `this.note` queries, media sources, and media-aware context suggestions.
- Added configurable HyDE lookup and improved ranking, source import, embedding saves, and index rebuild behavior.
#### v3.2.3
##### Use accelerated indexing with more embedding models
The accelerated index now supports embedding models that do not report their dimensions, including custom Ollama models. OpenAI embeddings also use the vector dimension you configure, so the index matches the model output you selected.
Re-imports and new embeddings now update the accelerated index in batches, reducing repeated index work as your vault changes.
##### See more of the index's memory footprint
Memory reporting now includes allocations used by the accelerated index in addition to loaded vectors.
*Embedding Health shows source coverage and vector memory together, including the accelerated index's own allocations.*
##### Full release notes
###### Indexing and model compatibility
- Batched accelerated-index updates and rebuilds after file re-imports and new embeddings.
- Source import and re-import queues now schedule block-index rebuilds.
- Fixed accelerated-index compatibility with models that do not report their embedding dimensions, including custom Ollama models.
- OpenAI embedding models now respect the configured vector dimension.
- Improved checks for whether a vector can be included in the accelerated index.
- Improved embedding initialization, unloading, and coordination with the vector index to clear old references and maintain consistent index state.
- Improved embedding saves.
###### Memory and exclusions
- Memory reporting includes WebAssembly allocations in addition to the loaded-vector usage reported by Core. Added access to the index's allocated buffer size.
- The file-exclusion dialog now supports pattern entry and .gitignore rules.
###### Shared infrastructure and maintenance
- Accelerated vector indexing and its Smart Environment v3 integration now belong to Pro. This includes sources, blocks, row-version tracking, external-vector projection, startup and shutdown, deletion maintenance, and release of JavaScript-held vectors.
- Added version-aware source and embedding management, separate in-memory and vector-index positions, and shared vector retrieval.
- Standardized source embedding-input preparation through shared actions and added Bases-specific embedding input.
- Added shared registration and invocation of Bases actions, with automated checks.
- Improved action registration and added automated checks that actions apply to the correct scope.
- Improved Environment event handling and added automated checks for Pro auto-updates.
- Added automated checks for source embedding input, vector-index behavior, vector retrieval, and embedding batches.
- Added a local model for discrepancy analysis. This is shared internal support, not a new discrepancy-analysis interface.
- Updated compatibility with Smart Environment v3.
---
## 4 8
canonical: https://smartconnections.app/smart-connections/releases/4-8/
html_url: https://smartconnections.app/smart-connections/releases/4-8/
markdown_url: https://smartconnections.app/smart-connections/releases/4-8.md
llms_url: https://smartconnections.app/smart-connections/releases/4-8/llms.txt
last_modified: 2026-09-10T21:02:27.715Z
usage_notes: |-
Use this page to answer questions about 4 8.
excerpt: |-
Smart Connections Pro v4.8 Change the model. Change the ranking. Keep the workflow. Smart Connections Pro is for vaults and workflows that have outgrown one default. v4.8 gives that advanced control a much stronger foundation: faster startup, more built-in local embedding models, model switching that preserves earlier embeddings, and visible scoring and ranking choices you can change without…
suggested_links:
- title: 4 5
url: https://smartconnections.app/smart-connections/releases/4-5/
- title: 4 6
url: https://smartconnections.app/smart-connections/releases/4-6/
- title: 4 7
url: https://smartconnections.app/smart-connections/releases/4-7/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
# Smart Connections Pro v4.8
## Change the model. Change the ranking. Keep the workflow.
Smart Connections Pro is for vaults and workflows that have outgrown one default. v4.8 gives that advanced control a much stronger foundation: faster startup, more built-in local embedding models, model switching that preserves earlier embeddings, and visible scoring and ranking choices you can change without rebuilding the entire setup.
The point is not more knobs. It is being able to adapt discovery to the vault, the language, and the task while keeping every decision inspectable and reversible.
> Update all installed Smart Plugins together, then restart Obsidian. Smart Connections Pro v4.8 requires Smart Environment v3.
## Try a different semantic model without losing the way back
The embedding model shapes what "related" means before Connections applies any result controls. Smart Environment v3 makes that choice much less destructive.
Choose from a broader built-in local catalog, including more lightweight and multilingual models. Activate another model without restarting Obsidian or deleting the embeddings created by the model you were using before. When you want to compare the results or return to the earlier setup, its embedding data is still there.
Smart Environment Pro also moves the accelerated index onto v3, with batched updates after re-imports and clearer memory reporting.
When tuning changes the list in an unexpected way, the improved Stats view and source inspector help you check whether the notes are embedded, skipped, or stale before blaming the ranking.
Learn more about the release of [Smart Environment v3](https://smartconnections.app/smart-environment/releases/3-0/).
## Choose how relevance is calculated
Connections Pro now puts the major stages of advanced result control within reach:
- **Filters** decide which notes are eligible to appear.
- **Scoring** measures the initial semantic fit.
- **Optional ranking** can reorder that candidate set using another strategy.
Scoring and ranking choices are available through visible menus and settings, so a specialized workflow does not have to depend on an invisible default. Change one stage, review the list, and revert when it does not improve the work.
## Filters now do what they say
Path filtering is more dependable, and **Exclude already linked** once again removes notes that are already connected to the target. That sounds small until a large vault fills the top of the list with material you have already handled.
Filters narrow the same Connections list you already know. Advanced control changes which notes enter the list and how they are ordered; it does not create a second workflow to learn.
## Turn the tuned result set into work
Once the list is useful, shared Smart Environment actions let it continue into Context, Graph, source menus, and supported Smart Tasks workflows. The result set does not have to be manually reconstructed just because the next step belongs in another plugin.
Pro also inherits the Core v4.7 target improvements: recent target history, block-level targeting inside the current note, file-drop retargeting, more consistent list menus, and configurable Footer Connections components.
## Before / After
| Before | With Smart Connections Pro v4.8 |
| --- | --- |
| Trying another embedding model felt like a one-way rebuild. | Switch models without deleting the embedding set you may want to return to. |
| The built-in local model choice was narrower. | Choose from more lightweight, multilingual, and experimental local options. |
| Advanced scoring and ranking behavior was harder to reach or explain. | Select the scoring and optional ranking stages from visible menus and settings. |
| Path and already-linked filters could admit unwanted candidates. | The filters once again narrow the list as configured. |
| Tuned results could become trapped in one surface. | Continue the reviewed set through shared Context, Graph, source, and task actions. |
## Supporting improvements
- Reusable source actions for listing connections, scoring one connection, and supported Smart Tasks workflows.
- Better source metadata and Context handling across advanced Connections actions.
- Better compatibility with Smart Environment v3 vectors and custom embedding dimensions.
- Commands and ribbon actions moved onto the shared Smart Environment action system.
## Learn more
- [Smart Connections overview](https://smartconnections.app/smart-connections/)
- [Smart Connections documentation](https://smartconnections.app/docs/connections/)
- [Smart Connections getting started](https://smartconnections.app/smart-connections/getting-started/)
- [Smart Connections FAQ](https://smartconnections.app/smart-connections/faq/)
## Release notes
### `v4.8.2`
#### Use passage-level Connections through Smart Connect Pro
Connections Pro now includes a tool action that Smart Connect Pro can expose to connected clients. Instead of returning one note-level list, it retrieves related sources for each embedded block separately, keeping results tied to the passage that prompted them.
Requests made through Smart Connect Pro use Connections Pro's own ranking, filtering, and scoring settings, helping the same Connections request behave consistently across the plugin interface and connected workflows.
Connections Pro provides the Connections action; Smart Connect Pro provides connector access to it.
#### See what re-ranking changed
Recency sorting now preserves the original similarity score, so you can still see how closely a result matched before its position changed. With a re-ranking model, enable **Show original connection score** to compare the original and re-ranked scores side by side.
*Keep the original similarity score visible so you can see what the re-ranking model changed.*
#### Full release notes
##### Smart Connect Pro actions
- Connections Pro now provides the `smart_connections_pro` tool action for Smart Connect Pro to expose to connected clients. It returns a separate Connections result list for each embedded block in a source.
- Added clearer guidance for requesting granular, per-block semantic results through the Connections Pro action.
- Requests made through Smart Connect Pro carry Connections Pro's local ranking, filtering, and scoring settings into the shared retrieval behavior.
##### Connections and ranking
- Updated the result-item display in the v4.2 Connections list to the v4 interface.
- Recency ranking preserves the underlying similarity score and displays how recently each result was edited.
- Improved re-ranking score handling and added **Show original connection score**, which displays the original score alongside the re-ranked score.
##### Maintenance
- Refactored inline Connections.
- Improved logic sharing between the Connections Pro interface and the tool actions used through Smart Connect Pro.
- Moved Connections-specific ranking, feedback scoring, and Connections-based Context suggestions from Smart Environment Pro into Smart Connections Pro so those features are owned by the product that provides them.
### `v4.8.1`
Updated: Smart Environment
### `v4.8.0`
Improved: is_vec detection should accept typed arrays for Smart Env v3 compatibility
Fixed: connections list path filtering
Fixed: filter should exclude already linked connections when setting is enabled
Added: ranking and scoring algorithm menu actions with corresponding configurations
migrated: ribbon icons to actions architecture
Migrate to action commands architecture
Add source connection actions and tests: implement list connections, score connection, and smart tasks functionalities
Enhance connection scoring and listing functionalities: update tests, improve metadata handling, and add source context support
Updated: Smart Environment v3
### v4.8.0
- Added visible menu actions and settings for scoring and optional ranking behavior.
- Fixed path filtering and **Exclude already linked** behavior.
- Added reusable source actions for listing connections, scoring a connection, and supported Smart Tasks workflows.
- Improved source metadata and Context handling across advanced Connections workflows.
- Unified Pro commands and ribbon actions on the Smart Environment v3 action system.
- Improved Smart Environment v3 vector compatibility.
- Inherited Smart Connections Core v4.7 target, handoff, menu, and display improvements.
- Updated Smart Connections Pro to Smart Environment Pro v3.2.0.
- Requires Smart Environment v3.0.0 or newer.
---
## 3 4
canonical: https://smartconnections.app/smart-context/releases/3-4/
html_url: https://smartconnections.app/smart-context/releases/3-4/
markdown_url: https://smartconnections.app/smart-context/releases/3-4.md
llms_url: https://smartconnections.app/smart-context/releases/3-4/llms.txt
last_modified: 2026-09-10T21:02:08.745Z
usage_notes: |-
Use this page to answer questions about 3 4.
excerpt: |-
Smart Context Pro v3.4 Build the full evidence set without losing control of it Real projects rarely fit in Markdown alone. Smart Context Pro extends the redesigned Builder with media, PDFs, embedded Bases, saved Contexts, and local rule overrides while keeping the complete source set visible before anything is copied or exported. More source types do not have to mean a more mysterious payload.…
suggested_links:
- title: 3 1
url: https://smartconnections.app/smart-context/releases/3-1/
- title: 3 2
url: https://smartconnections.app/smart-context/releases/3-2/
- title: 3 3
url: https://smartconnections.app/smart-context/releases/3-3/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
# Smart Context Pro v3.4
## Build the full evidence set without losing control of it
Real projects rarely fit in Markdown alone. Smart Context Pro extends the redesigned Builder with media, PDFs, embedded Bases, saved Contexts, and local rule overrides while keeping the complete source set visible before anything is copied or exported.
More source types do not have to mean a more mysterious payload. Build first, inspect second, then choose exactly how the package should leave Obsidian.
> Update all installed Smart Plugins together, then restart Obsidian. Smart Context Pro v3.4 requires Smart Environment v3.
## One Builder for mixed evidence
Review notes, sections, named Contexts, images, PDFs, folders, and supported embedded Bases in the same tree. The source count, token estimate, rule count, and per-source contribution remain visible even when the project grows beyond ordinary Markdown.
Media-aware Context codeblocks and menus use that same Builder state. You no longer need one mental model for notes and another for the files that travel with them.
## Rules no longer hide in settings
The Builder now shows the include and exclude rules that shape one Context. See which folder or saved Context introduced a source, restore an exact exclusion, or turn off a broader global exclusion only for this package.
The exception stays local. The wider folder, named-context, or global rule continues protecting the rest of the vault, so one deliberate override does not become future cleanup work.
## Choose the output, not just the sources
**Copy text** and **Copy media** are now separate actions. A text prompt no longer shares one ambiguous clipboard operation with images and PDFs.
- Copy the reviewed text from the Builder, a Context codeblock, or supported menus.
- Choose the supported media files that should travel with the current Context depth.
- Export the full package as a ZIP when the destination needs files rather than clipboard content.
- See progress and clearer handling when a ZIP is too large to complete normally.
## Embedded Bases stay grounded in their source
Supported Bases views can render as part of the Context package, including relative `this.file` and `this.note` queries. The Builder can keep that output tied to the note that gives the query meaning, and a missing Bases file no longer has to derail the rest of the package.
You can also decide whether embedded Bases render inline, keeping a dense package readable without removing the source entirely.
## Smart Environment v3 supplies the stronger rails
Context Pro inherits v3's faster suite startup, broader built-in local embedding-model catalog, and non-destructive model switching. Drag-and-drop preserves the source you moved, and source menus behave more consistently across the suite.
The improved Environment inspector also makes it easier to verify what was imported or embedded when a semantic source workflow behaves unexpectedly.
Supported Pro integrations can now keep API keys in Obsidian's native keychain-backed secret storage instead of ordinary plugin settings.
Learn more about the release of [Smart Environment v3](https://smartconnections.app/smart-environment/releases/3-0/).
## Before / After
| Before | With Smart Context Pro v3.4 |
| --- | --- |
| Mixed notes and media became difficult to inspect as one package. | The redesigned Builder keeps richer source types in one visible tree. |
| Text and media relied on one copy path with destination-dependent behavior. | Choose explicit text, media, or ZIP outputs after reviewing the source set. |
| Dynamic Bases could lose the source needed for relative queries. | Embedded Bases resolve from the Context source, including `this.file` and `this.note`. |
| One local exception could tempt you to weaken a global exclusion. | Override the rule for one Context while preserving broader defaults. |
| Large, file-heavy context had to fit through the clipboard. | Export a ZIP with visible progress and safer oversized-output handling. |
## Supporting improvements
- Media-aware Context codeblocks, commands, ribbon actions, and menus.
- Better handling for missing Bases files and source-relative Base compilation.
- More resilient rule restoration across folders and included named Contexts.
- Fixed Context codeblocks whose names contain a plus sign.
- Shared Smart Environment v3 actions across file menus, commands, ribbons, and codeblocks.
## Learn more
- [Smart Context overview](https://smartconnections.app/smart-context/)
- [Smart Context documentation](https://smartconnections.app/docs/context/)
- [Smart Context getting started](https://smartconnections.app/smart-context/getting-started/)
- [Smart Context FAQ](https://smartconnections.app/smart-context/faq/)
## Release notes
### `v3.4.2`
#### Bring Base views directly into Context
Use the results of an individual Base view alongside notes in the same Context. Views render on demand as Markdown tables and can resolve queries relative to the current note. Different views of the same Base stay distinct, so each can contribute its own results and links.
#### Add images alongside your notes
Images can now join Context as Smart Sources and appear in Builder suggestions. Before copying, the media chooser shows how much media was detected so you can decide whether to include it.
*Check the detected media count and total size before deciding whether to include it.*
#### Use Context Pro through Smart Connect Pro
Context Pro now includes tool actions that Smart Connect Pro can expose through its supported connector channels. With both plugins installed, connected clients can create and list named Contexts, update descriptions, add or remove sources, and retrieve a Context as a manifest, compiled text, file tree, or ZIP data.
You can also export the registered Smart Sources in a folder without first creating a named Context. External-folder scans apply your active exclusions before counting files toward the scan limit.
Context Pro provides the Context actions; Smart Connect Pro provides access to them through Obsidian CLI, Local MCP, or both.
*In Smart Connect Pro, choose whether Context Pro actions are available through Obsidian CLI, Local MCP, or both.*
#### Full release notes
##### Bases and media
- Added a Bases view type that makes its results available as context items.
- Individual Bases views render on demand when read as sources.
- Bases views support Markdown table output and relative queries, including `this.file` and `this.note`.
- Each Bases view stores its own links and embeddings, without a separate Bases cache collection.
- Previously imported Bases blocks can be repaired when view indexes or block metadata are incomplete.
- Links in context-dependent Bases views resolve without changing the indexed Base data.
- Improved reading and rendering of embedded Bases.
- Images can now be registered as Smart Sources, with media suggestions available when building context.
##### Smart Connect Pro actions
- Context Pro now provides the Context-management tool actions that Smart Connect Pro can expose through its connector channels.
- With Smart Connect Pro, connected clients can create, retrieve, and list named Contexts, and add or remove their included sources without editing the source notes.
- Add a description when creating a Context, update it later through `context_update`, and read descriptions in Context listings.
- When two names resolve to the same saved Context identity, the new Context receives a different name. The tool response includes the requested name and whether it was renamed.
- External-folder scans apply active exclusions before counting files toward the scan limit.
##### Retrieval and export
- Context retrieval through Smart Connect Pro supports a manifest, compiled text, a file tree, or ZIP output.
- Export the registered Smart Sources in a folder as a file tree or ZIP without first creating a named Context.
- ZIPs returned through tool actions are built in memory and delivered as base64-encoded data; they are not automatically saved to disk.
##### Compatibility and maintenance
- The deprecated `context_read` action has been removed. Integrations using Context Pro through Smart Connect Pro should use `context_get`, exposed as `smart_context_get`.
- Context Pro tool actions include clearer descriptions and usage guidance.
- Expanded automated checks for Context management and export.
### `v3.4.1`
Updated: Smart Environment
### `v3.4.0`
Enhanced: context codeblock should support media files
Improved: removed the Copy media by default setting and replaced it with an explicit Copy current media command.
Improved: Smart Context copy UX now separates Copy text and Copy media actions instead of presenting media as a default copy mode to resolve mixed clipboard payloads pasting inconsistently across destinations..
Added: ribbon icon for copying media from current context to clipboard
Improved: enhance embedded base rendering by adding a toggle in settings to control inline rendering of embedded bases in source items.
Added: context renders embedded bases with relative (`this.`) queries
Added: implement media copy functionality with context menu integration
Fixed: should handle missing bases files when rendering bases embeds
Improved: file-nav menu options should use menu action pattern
Added: implement context export as ZIP file functionality
Improved: embedded base file handling when compiling context
Migrated: commands and ribbon icons to command actions architecture
Resolved: codeblock failing to work with context names that include plus sign
removed dependency on data.blocks to support v3 transition
Added: copy ribbon icon now opens menu
Improved: New v2 context builder UI
feat: implement rule management for SmartContext
- Add `rules_list.js` to handle dynamic include and exclude rules for contexts.
- Introduce functions to restore exclusions, open contexts, and group rule entries.
- Enhance context exclusion utilities to support rule entries and improve data structure.
- Update tests to cover new functionalities for context rules and exclusions.
Improved: ZIP export functionality with oversized handling and progress notifications
Improved: source_get_context for better handling bases
Improved: copy ribbon icon menu actions with sub-menus for copy types
Added: override global-level exclusion settings inside individual context rules
Refactor SmartContext exclusion handling and enhance tests
- Improved exclusion pattern retrieval in SmartContext to include selected named contexts and normalize exclusion keys.
- Added functionality to restore local exclusions when adding items, ensuring broader rules remain intact.
- Enhanced removal logic to handle surviving folder and named context rules without hydration.
- Introduced new utility functions for explicit exclusion identity and normalization of exclusion match keys.
- Updated context exclusion tests to cover new behaviors and edge cases, ensuring robust handling of exclusions.
Updated: Smart Environment v3
Refactor exclusion handling across context and item management
### v3.4.0
- Extended the redesigned Context Builder with Pro media, Bases, rule, and export workflows.
- Added explicit **Copy text** and **Copy media** actions across the Builder, codeblocks, commands, ribbons, and supported menus.
- Added media-aware Context codeblocks and a dedicated media chooser.
- Added embedded Bases rendering, relative `this.file` and `this.note` queries, inline-rendering control, and safer missing-file handling.
- Added per-context exclusion overrides and improved rule restoration across folders and named Contexts.
- Added ZIP export with progress and oversized-output handling.
- Fixed Context codeblock names containing a plus sign.
- Updated Smart Context Pro to Smart Environment Pro v3.2.0.
- Requires Smart Environment v3.0.0 or newer.
---
## Custom Algorithms
canonical: https://smartconnections.app/smart-connections/custom-algorithms/
html_url: https://smartconnections.app/smart-connections/custom-algorithms/
markdown_url: https://smartconnections.app/smart-connections/custom-algorithms.md
llms_url: https://smartconnections.app/smart-connections/custom-algorithms/llms.txt
last_modified: 2026-09-10T19:30:49.636Z
usage_notes: |-
Use this page to answer questions about Custom Algorithms.
excerpt: |-
Tune Smart Connections scoring and ranking algorithms Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Connections scoring documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. Use this page after the default Connections workflow already works, but a recurring result…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
- title: Inline
url: https://smartconnections.app/smart-connections/inline/
# Tune Smart Connections scoring and ranking algorithms
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Connections scoring documentation](https://smartconnections.app/docs/connections/#smart-connections-scoring) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
Use this page after the default Connections workflow already works, but a recurring result problem remains.
For example:
- useful notes appear, but the order is consistently wrong
- results come from too broad a candidate pool
- a project, folder, tag, or frontmatter pattern should matter more
- the same low-value notes keep returning
- a recurring workflow needs a repeatable ranking strategy
Connections Pro exposes scoring and ranking controls for those result-shaping problems.
If you have not seen any useful Connections result yet, start with [Getting Started](https://smartconnections.app/smart-connections/getting-started/) instead of tuning algorithms.
Start with the symptom:
| Symptom | Tune first | Why |
| --- | --- | --- |
| Results are too broad | Results type, limits, filters | The candidate pool is probably too wide. |
| Results are relevant but ordered poorly | Ranking algorithm | The score is acceptable, but final order needs shaping. |
| Valuable project, folder, tag, or frontmatter notes are under-ranked | Score algorithm weights | The relevance signal needs metadata or path emphasis. |
| Same low-value notes keep returning | Feedback-aware scoring or filters | You need to steer recurring noise. |
| Similar notes are actually repeated work | Smart Dedupe | Cleanup is a review decision, not a ranking tweak. |
## Tuning order
Recommended order:
1. Choose the candidate pool.
2. Adjust filters and limits.
3. Select a score algorithm.
4. Add ranking only when you need extra shaping.
5. Use Dedupe when repeated material needs review, not when the list merely needs tuning.
Change one layer at a time. If you change scope, score, and ranking together, it becomes hard to know what helped.
## Candidate pool
Choose a results collection key.
- `Smart Sources` for note-level discovery.
- `Smart Blocks` for heading/block-level precision.
Use Sources when you want broader note-level overview.
Use Blocks when long notes hide the useful section and you need finer-grained matches.
If Blocks results feel noisy or heavy, return to Sources or tune block settings in Smart Environment.
## Filters before algorithms
Filters decide what is allowed into the list.
Use filters before advanced scoring when the problem is scope.
Good filter use cases:
- keep results inside a client, project, area, or folder
- hide archives, templates, exports, or completed work
- focus on frontmatter such as `status:open` or `type:meeting`
- remove known low-value result pools before scoring
If you want notes excluded from indexing entirely, use Smart Environment exclusions instead.
Connections filters affect what is shown after Smart Environment has prepared the data.
Related:
- [Smart Connections settings](https://smartconnections.app/smart-connections/settings/)
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
## Score algorithms
Score algorithms decide the primary relevance score for each candidate.
For tuning comparisons, use `Cosine Similarity` as the baseline.
Move to feedback or metadata weighting after you observe noisy, under-weighted, or over-weighted results.
| Algorithm | What it does | When to use |
| --- | --- | --- |
| Cosine Similarity | Ranks by embedding similarity between the current note and candidates. | Baseline when you want stable, feedback-free results. |
| Similarity Adjusted by Feedback | Penalizes candidates similar to hidden notes. | When recurring noise keeps returning after you hide it. |
| Similarity Weighted by Feedback | Boosts candidates similar to pinned notes and dampens hidden notes. | When pinned and hidden signals represent real preference for a repeatable workflow. |
| Similarity Weighted by Key + Frontmatter | Multiplies similarity based on key fragments and frontmatter matches. | When metadata, paths, or headings should emphasize or de-emphasize results. |
## Ranking algorithms
Ranking algorithms reorder already-scored candidates.
Use ranking when results are relevant but the final order feels wrong.
| Algorithm | What it does | Notes |
| ---------------- | ----------------------------------------------------- | ------------------------------------------------------------------------- |
| None | Keeps the original score order. | Fastest option and best baseline for comparison. |
| Re-ranking model | Applies a reranking model to reorder the scored list. | Requires a configured Smart Rank model in Smart Environment Pro settings. |
| Recency rank | Orders results by most recently modified items. | Useful when freshness should dominate final order. |
### base example for comparison reference point
### re-ranking model example
### recency example
## Score vs ranking boundary
Keep the boundary simple:
- **Score algorithms** decide relevance.
- **Ranking algorithms** decide final order after scoring.
- **Filters** decide which candidates are allowed into the list.
A practical rule:
> Fix scope first. Tune relevance second. Reorder third. Clean up repeated work only after review.
## Controls at a glance
- Results collection key selects Smart Sources or Smart Blocks as the candidate pool.
- Filters narrow which candidates appear.
- Score algorithm selects the primary relevance strategy.
- Ranking algorithm applies a secondary ordering stage.
- Post processing can run a configured rerank stage when a ranking model is available.
## Practical presets
### Default, low-maintenance
- Smart Sources
- Cosine Similarity
- None
Use this when you want stable, broad note-level discovery.
### Reduce recurring noise
- Smart Sources
- Similarity Adjusted by Feedback
- None
Use this after hiding notes that keep polluting useful results.
### Leverage explicit curation
- Smart Sources
- Similarity Weighted by Feedback
- Re-ranking model if needed
Use this when pinned and hidden signals represent real preference.
### Metadata-driven retrieval
- Smart Blocks
- Similarity Weighted by Key + Frontmatter
- Recency rank if freshness matters
Use this when folders, keys, headings, or frontmatter should shape relevance.
## Example weighting config
Use JSON in the score algorithm settings:
```json
{
"key_weights": {
"Projects/": 1.2,
"Readwise/": 0.8,
"#High-value heading": 1.3
},
"meta_weights": {
"status:evergreen": 1.15,
"type=spec": 1.1,
"reviewed": 1.05
}
}
```
- `key_weights` applies substring matches to key, path, or heading signals.
- `meta_weights` supports `key`, `key:value`, and `key=value` matchers.
- Multiple matches multiply together.
This is best for structural intent.
Use recency ranking for freshness intent.
## Tuning workflow
### 1. Establish a baseline
Start with:
- Smart Sources
- Cosine Similarity
- no ranking
- minimal filters
Open one real note and write down what feels wrong.
Do not tune against an empty test note or a note that has not produced any useful Connections result yet.
### 2. Fix scope
If results come from the wrong part of the vault, use filters first.
Do not use a ranking model to solve a scope problem.
### 3. Fix granularity
If whole notes are too broad, test Smart Blocks.
If blocks are too noisy or heavy, return to Sources or tune block settings in Smart Environment.
### 4. Fix signal
If the right notes exist but are consistently under-weighted, add metadata or feedback weighting.
Keep the candidate pool stable while testing signal changes.
### 5. Fix final order
If the right candidates are present but poorly ordered, add ranking.
Compare against the baseline before keeping the change.
### 6. Use Dedupe for repeated work
If the problem is that similar blocks should be compared, merged manually, archived, or ignored, use [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/).
Similarity creates the question.
Dedupe review turns it into a decision.
## Troubleshooting quick checks
### Results feel too broad
Try:
- lowering result limits
- adding include filters
- excluding archives, templates, or exports
- switching to Smart Blocks when long notes hide the useful section
Use [Connections settings](https://smartconnections.app/smart-connections/settings/) for the exact controls.
### Same low-value notes keep returning
Try:
- hiding those notes in the Connections workflow
- applying feedback-adjusted scoring
- adding path or frontmatter excludes
- confirming they should not be excluded in Smart Environment instead
### Valuable tagged or foldered notes are under-ranked
Try:
- configuring key/frontmatter weights
- checking that metadata is spelled consistently
- keeping the score algorithm stable while testing weights
### Results are relevant but ordering feels off
Try:
- adding a ranking algorithm
- testing a reranking model
- using recency rank when freshness matters
- comparing against the baseline before keeping the change
### Similar results look like duplicates
Use Dedupe, not ranking.
Connections algorithms decide retrieval order.
Dedupe helps review likely repeated blocks or notes side by side.
### No useful results appear at all
Do not start with algorithm tuning.
Use [Getting Started](https://smartconnections.app/smart-connections/getting-started/) to verify note eligibility and vault coverage first.
## Related pages
- [Explore the Smart Connections view](https://smartconnections.app/smart-connections/list-feature/)
- [Tune Smart Connections settings](https://smartconnections.app/smart-connections/settings/)
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
- [Connections in Bases](https://smartconnections.app/smart-connections/bases/)
- [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/)
---
## Smart Connections Getting Started
canonical: https://smartconnections.app/story/smart-connections-getting-started/
html_url: https://smartconnections.app/story/smart-connections-getting-started/
markdown_url: https://smartconnections.app/story/smart-connections-getting-started.md
llms_url: https://smartconnections.app/story/smart-connections-getting-started/llms.txt
last_modified: 2026-09-10T14:04:14.660Z
usage_notes: |-
Use this page to answer questions about Smart Connections Getting Started.
excerpt: |-
Welcome to Smart Connections Open guide Notes everywhere. Progress nowhere? Useful ideas hide behind folders, titles, and old wording Manual linking and rereading steals writing time Start from the note in front of you Open a project, meeting, draft, research, or decision note The note in view becomes the anchor No query or prior links required Open guide Open Connections Run: Smart Connections:…
suggested_links:
- title: Build Moc With Smart Connections
url: https://smartconnections.app/story/build-moc-with-smart-connections/
- title: Smart Chat Getting Started
url: https://smartconnections.app/story/smart-chat-getting-started/
- title: Smart Connect Getting Started
url: https://smartconnections.app/story/smart-connect-getting-started/
- title: Smart Connections Getting Started New
url: https://smartconnections.app/story/smart-connections-getting-started-new/
- title: Smart Connections
url: https://smartconnections.app/story/smart-connections/
# Welcome to Smart Connections
[Open guide](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-welcome)
# Notes everywhere. Progress nowhere?
- Useful ideas hide behind folders, titles, and old wording
- Manual linking and rereading steals writing time
# Start from the note in front of you
- Open a project, meeting, draft, research, or decision note
- The note in view becomes the anchor
- No query or prior links required
[Open guide](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-start-real-note)
# Open Connections
- Run: Smart Connections: Open connections view
- Or click the connections icon
- Keep it open in the sidebar while you work
[Open steps](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-open-connections)
# Create your first related link
- Preview one promising result
- Drag it into the note to create a link
- Keep only the relationship that helps the work
[Create links](https://smartconnections.app/smart-connections/list-feature/?utm_source=sc-story-first-link-drag#core-moves)
# You know it worked when...
- A related note changes what you write next
- The current note gains an insight, source, decision, or link
- Progress in the note is the value
[Finish quick path](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-know-it-worked)
"I moved my entire workflow to Obsidian because of this plugin."- Community member
"Tamed decades of scattered research in one evening."- Ronny
"Cuts case-law summarizing time in half."- Carey
# The current note is the anchor
- Results update as you switch notes
- Score is a lead to inspect, not a grade
- Pause holds one anchor steady
[List controls](https://smartconnections.app/smart-connections/list-feature/?utm_source=sc-story-current-note-anchor#understanding-ui)
# Weak results? Use a richer note
- Empty or tiny notes provide weak anchors
- Open a note with project, research, meeting, or decision text
- Try Connections again
[Fix missing results](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-richer-note#if-useful-results-do-not-appear)
# Let preparation finish
- Smart Environment prepares eligible notes for local retrieval
- No API key is needed for Core retrieval
- Readiness depends on vault size and device
[Check readiness](https://smartconnections.app/smart-environment/milestones/?utm_source=sc-story-readiness)
# Still empty? Inspect the note
- Use the Smart Environment status menu
- Run: Inspect active note
- Check that should embed and vectorized are green
[Inspect note](https://smartconnections.app/smart-environment/settings/?utm_source=sc-story-inspect-active-note)
# Check vault coverage
- Run: Show stats
- Embedding coverage should be greater than zero
- Use Milestones when preparation is still running
[Check coverage](https://smartconnections.app/smart-environment/milestones/?utm_source=sc-story-check-vault-coverage)
# Watch related notes appear
- See a note become a stronger anchor as it gains meaning
- Notice how related notes change without a search query
[Watch segment](https://www.youtube.com/watch?v=_i3577ti8jg&t=116s&utm_source=sc-story-video-segment)
# Make it routine
- Leave Connections open beside your notes
- When the note makes you search, preview one result
- Apply one useful result before moving on
[Make routine](https://smartconnections.app/smart-connections/list-feature/?utm_source=sc-story-make-it-routine)
# Need daily list controls?
- Pause one anchor while browsing
- Copy a clean list of links
- Hide noise or pin references
[Use list view](https://smartconnections.app/smart-connections/list-feature/?utm_source=sc-story-daily-list-controls)
# Full notes or precise excerpts?
- Sources return whole notes for overview
- Blocks return smaller sections for precision
- Start broad, switch when the useful part is buried
[Read Blocks FAQ](https://smartconnections.app/smart-connections/faq/?utm_source=sc-story-blocks-sources#what-does-sources-vs-blocks-mean)
# Useful but noisy?
- Lower result limit for clarity
- Move the sidebar where it fits your layout
- Tune after one useful result, not before
[Tune settings](https://smartconnections.app/smart-connections/settings/?utm_source=sc-story-settings-tune)
# Finish notes with related links
- Footer connections appear at the end of the note
- Useful on mobile and no-sidebar layouts
- Review related notes before leaving the page
[Use Footer links](https://smartconnections.app/smart-connections/footer/?utm_source=sc-story-footer-connections)
# Starting from a question?
- Ask in plain language
- Search by meaning, not exact keywords
- Use Lookup when the query is the anchor
[Explore Lookup](https://smartconnections.app/smart-lookup/search/?utm_source=sc-story-lookup-question)
# Need paragraph-level context?
- Inline connections appear beside specific blocks
- Hover to scan related matches in the editor
- Use after the whole-note workflow works
[Learn Inline](https://smartconnections.app/smart-connections/inline/?utm_source=sc-story-inline-connections)
# Open the exact match
- Click a block match to open the note
- Use this when a paragraph needs closer review
- Return to the current paragraph with the useful part
[Learn Inline](https://smartconnections.app/smart-connections/inline/?utm_source=sc-story-inline-open-note)
# Preview without leaving flow
- Hold Cmd or Ctrl while hovering
- Inspect the related block before switching context
- Keep only the match that helps the sentence or paragraph
[Learn Inline](https://smartconnections.app/smart-connections/inline/?utm_source=sc-story-inline-hover-preview)
# Need relevance in a table?
- Add a Connections score column to a Base
- Compare rows against a reference note
- Sort planning, research, or triage by relevance
[Explore Bases](https://smartconnections.app/smart-connections/bases/?utm_source=sc-story-bases-score-column)
# Want link trails per row?
- Use list_connections in a Bases column
- See related links beside each row
- Build review dashboards without leaving Bases
[Use Bases links](https://smartconnections.app/smart-connections/bases/?utm_source=sc-story-bases-link-trails#formulas)
# Recurring result problem?
- Fix scope first
- Then tune scoring or ranking
- Change one layer at a time
[Tune ranking](https://smartconnections.app/smart-connections/custom-algorithms/?utm_source=sc-story-custom-algorithms)
# Need the shape of a topic?
- Use a graph when a list feels too flat
- Look for clusters, bridges, and neighborhoods
- Act on selected notes after the shape is useful
[Explore Graph](https://smartconnections.app/smart-graph/?utm_source=sc-story-smart-graph)
# Turning matches into AI context?
- Use Connections to find strong candidates
- Review what belongs before packaging
- Send selected material when the next step needs AI
[Build context](https://smartconnections.app/smart-context/clipboard/?utm_source=sc-story-smart-context)
# Still blocked?
- Check the FAQ before changing everything
- Use the page that matches your symptom
- Then return to one meaningful note
[Open FAQ](https://smartconnections.app/smart-connections/faq/?utm_source=sc-story-faq-support)
# Get one useful related note today
- Open a meaningful note
- Open Connections
- Preview and apply one result
[Learn Smart Loop](https://smartconnections.app/smart-loop/?utm_source=sc-story-final-cta)
[Start onboarding](https://smartconnections.app/onboarding/start/?utm_source=sc-story-final-cta)
# Watch Wanderloots' Walkthrough
See Smart Connections in action.
[Start onboarding](https://smartconnections.app/onboarding/start/?utm_source=sc-story-callum-video) | [Smart Loop](https://smartconnections.app/smart-loop/?utm_source=sc-story-callum-video)
[Watch on YouTube](https://www.youtube.com/watch?v=7Rvl9Sl29Jk)
---
## Smart Connections Getting Started New
canonical: https://smartconnections.app/story/smart-connections-getting-started-new/
html_url: https://smartconnections.app/story/smart-connections-getting-started-new/
markdown_url: https://smartconnections.app/story/smart-connections-getting-started-new.md
llms_url: https://smartconnections.app/story/smart-connections-getting-started-new/llms.txt
last_modified: 2026-09-10T14:03:49.796Z
usage_notes: |-
Use this page to answer questions about Smart Connections Getting Started New.
excerpt: |-
I know I wrote this somewhere. Start from the note already in front of you. Open guide Open Connections. Run Open: Connections view. Core starts with Smart Connections. Pro starts with Smart Connections Pro. Open steps Related notes. No search required. Connections ranks them from the note already open. Read result rows Read the part that matched. Expand a result without leaving the note. Expand…
suggested_links:
- title: Build Moc With Smart Connections
url: https://smartconnections.app/story/build-moc-with-smart-connections/
- title: Smart Chat Getting Started
url: https://smartconnections.app/story/smart-chat-getting-started/
- title: Smart Connect Getting Started
url: https://smartconnections.app/story/smart-connect-getting-started/
- title: Smart Connections Getting Started
url: https://smartconnections.app/story/smart-connections-getting-started/
- title: Smart Connections
url: https://smartconnections.app/story/smart-connections/
# I know I wrote this somewhere.
Start from the note already in front of you.
[Open guide](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-welcome)
# Open Connections.
- Run **Open: Connections view**.
- Core starts with Smart Connections. Pro starts with Smart Connections Pro.
[Open steps](https://smartconnections.app/docs/connections/?utm_source=sc-story-open-connections#connections-view-open-view)
# Related notes. No search required.
Connections ranks them from the note already open.
[Read result rows](https://smartconnections.app/docs/connections/?utm_source=sc-story-see-what-surfaces#connections-view-result-rows)
# Read the part that matched.
Expand a result without leaving the note.
[Expand a result](https://smartconnections.app/docs/connections/?utm_source=sc-story-see-the-source#connections-view-expand-collapse)
# You know it worked when...
You can say what this source changes in the work.
[Use what you found](https://smartconnections.app/smart-connections/getting-started/?utm_source=sc-story-know-it-worked#use-the-result)
# Have a question instead?
Type the idea. Smart Lookup finds related sources.
[Try Smart Lookup](https://smartconnections.app/smart-lookup/getting-started/?utm_source=sc-story-lookup-question)
# Need these notes for AI?
Send the set to Smart Context, then remove what does not belong.
[Build context](https://smartconnections.app/smart-context/getting-started/?utm_source=sc-story-smart-context)
# Need the shape, not another list?
Smart Graph lets you explore several related notes as a visual neighborhood.
[Explore Smart Graph](https://smartconnections.app/smart-graph/getting-started/?utm_source=sc-story-smart-graph)
# Only add another plugin when the job changes.
Templates, Chat, Dedupe, and Connect Pro each have their own first win.
[Explore Smart Plugins](https://smartconnections.app/getting-started/?utm_source=sc-story-more-smart-plugins)
[Watch segment](https://www.youtube.com/watch?v=_i3577ti8jg&t=116s&utm_source=sc-story-video-discovery-layer)
[Watch on YouTube](https://www.youtube.com/watch?v=7Rvl9Sl29Jk&utm_source=sc-story-callum-video)
---
## 2 0
canonical: https://smartconnections.app/smart-connect/releases/2-0/
html_url: https://smartconnections.app/smart-connect/releases/2-0/
markdown_url: https://smartconnections.app/smart-connect/releases/2-0.md
llms_url: https://smartconnections.app/smart-connect/releases/2-0/llms.txt
last_modified: 2026-09-08T23:48:20.395Z
usage_notes: |-
Use this page to answer questions about 2 0.
excerpt: |-
Smart Connect Pro v2.0 Delegate changes to AI - review them before they touch your notes Connect Pro can now turn corrections and reorganizations suggested by ChatGPT, Claude, or another connected client into reviewable changes inside Obsidian. Instead of choosing between manual copy/paste and letting an agent edit immediately, you get a third option: let it prepare the change, inspect exactly…
suggested_links:
- title: 1 0
url: https://smartconnections.app/smart-connect/releases/1-0/
- title: 1 2
url: https://smartconnections.app/smart-connect/releases/1-2/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
# Smart Connect Pro `v2.0`
## Delegate changes to AI - review them before they touch your notes
Connect Pro can now turn corrections and reorganizations suggested by ChatGPT, Claude, or another connected client into reviewable changes inside Obsidian.
Instead of choosing between manual copy/paste and letting an agent edit immediately, you get a third option: let it prepare the change, inspect exactly what will happen, adjust it if needed, then Apply or Reject it yourself.
The new **Connect Pro Inbox** makes delegated vault work practical while keeping the final decision in Obsidian.
*Corrections, block moves, and excerpt moves share one Inbox. All three items shown are **Pending**, with a separate **Review** action for each.*
### Let agents prepare real vault changes
Connected agents can now propose three kinds of changes without immediately editing your source notes:
- **Correct a claim** - replace one exact passage using supporting source evidence.
- **Move an excerpt** - relocate selected text, with optional revised wording at the destination.
- **Move blocks** - reorganize complete sections, including their child sections, in an explicit order.
Open the Inbox from the Connect Pro ribbon icon, command palette, or Connect Pro settings.
Pending proposals remain there until you decide what to do with them. Priority controls review order, while Group and Label make larger batches easier to navigate. Each proposal still has its own decision; a group is not a shared approval.
### Inspect and refine the exact result
Corrections and excerpt moves include **Changes**, **Before**, and **After** views so you can inspect the resulting files before applying anything. A new destination has no **Before** view because the file does not exist yet.
You can edit the proposed replacement or insertion directly in the review, change the destination of a move, or reorder selected blocks. **Done editing** keeps a review draft without changing source notes; applying the proposal is a separate decision.
***Done editing** keeps the proposed replacement as a review draft; it does not apply the correction. Check the evidence and refreshed diff before using the separate **Apply correction** action.*
For structural moves, Connect Pro shows the selected content, included child sections, resulting heading placement, and exact file changes.
If the source changes while you are reviewing it, inspect a fresh preview before applying. If the exact excerpt or selected block no longer matches, read the current source and make a new proposal.
### Bring selected sections together without merging whole notes
In this separate captured example, **Reader need** from **Reader interviews** and **Weekly format** from **Editorial plan** are proposed for the existing **Launch playbook**. The selected text and heading levels remain unchanged; the destination and order are explicit.
*Initial pending order: **Reader need**, then **Weekly format**. **Up** beside Weekly format changes that order; this preview does not show an applied move.*
The reviewer then moves **Weekly format** ahead of **Reader need**, checks both removals in **Source changes**, and applies **Move 2 blocks**. The saved result below therefore follows the revised order, not the initial order shown above.
*Saved result after review and application: **Launch playbook** keeps its original **Audience** section and gains **Weekly format**, then **Reader need**. Both source notes remain with their unselected sections.*
This is selected-section movement, not generated synthesis, copying, or deletion of the original notes.
### Stay in control: Apply, Reject, or come back later
**Apply** performs the reviewed change after checking the current source again.
**Reject** records the decision without changing the source files.
**Skip** leaves the proposal pending so you can return to it later.
Applied and rejected proposals remain available in **History**. History records decisions; inspect the actual affected notes to verify the resulting contents.
When an excerpt is both moved and rewritten, confirm the original removal and the rewritten insertion separately before **Move and rewrite** becomes available. A relevant edit requires a fresh preview and resets those confirmations.
*Both confirmations are unchecked, so **Move and rewrite** is disabled. Review the rewritten insertion and exact original removal separately. **Link impact not checked** remains an explicit limit.*
### Give future agents useful review history
The read-only `get_proposal_feedback` Tool lets connected agents retrieve saved proposals and review decisions.
Connected agents can inspect previous proposals and review outcomes, including available comparisons between proposed and applied values, to understand relevant prior decisions. Missing original comparison evidence is not proof that a proposal was applied unchanged.
This history is evidence, not an automatic preference system. Agents should still inspect current notes and follow your current instructions before proposing another change.
### Use the same Smart Tools from MCP and CLI
Local MCP makes configured Smart Tools from your running Obsidian vault available to supported desktop clients.
Smart CLI Pro, or Smart CLI, uses the same underlying Smart Tool contracts for compatible command-line workflows. Calling a proposal Tool through CLI still creates a pending Inbox item; it does not bypass review.
Connect Pro settings make it easier to see which capabilities are available and which connector they are enabled for.
Local MCP runs independently of the legacy Remote GPT connection, so a desktop MCP client does not require that route to be connected.
### Safer when the vault changes underneath you
This release includes substantial reliability work behind the Inbox.
Connect Pro rechecks the reviewed plan against current source content before applying it, detects stale reviews, verifies written changes, and attempts conservative recovery when an application fails partway through.
If file changes or the saved review decision remain unresolved, **Manual recovery required** blocks the affected proposal from application. Inspect the named files and saved review state before proposing another change. Inspection alone does not reset the block, and rejecting the item does not restore partially changed content.
When a change is already applied but source indexing needs a refresh, inspect the files and refresh indexing rather than applying the change again.
### Try the propose-review-apply workflow
Connect a supported desktop client through Local MCP, refresh its available Tools, and start with one recognizable read from your vault.
Then use a disposable note for one small `propose_correction`, `propose_move_excerpt`, or `propose_move_blocks` request. A successful proposal returns `status: pending` and `source_files_changed: false`.
Check that the source note is unchanged, review the proposal in the Inbox, then Apply or Reject it. If you apply it, compare the actual affected files with the reviewed result. Pending is the expected proposal outcome, not a missing write to retry.
### Know what is - and is not - held for review
Inbox review applies to the dedicated proposal actions. It does not automatically intercept every Tool or native Obsidian CLI write. Review availability for the channel you use; individual native-CLI controls do not restrict the broad `obsidian_cli` Tool.
Connect Pro does not currently provide automatic Undo, automatic link repair, or crash-atomic multi-file transactions.
Local MCP connects to the running vault on the same computer. The client you connect may still send returned content to its model provider.
If you used earlier experimental proposal workflows, use the current `propose_correction`, `propose_move_excerpt`, and `propose_move_blocks` names. Older separate contradiction, Extract, and Merge stores are not imported into the unified Inbox.
## Learn more
- [Getting started](https://smartconnections.app/connect-pro/getting-started/)
- [Documentation](https://smartconnections.app/docs/connect-pro/)
- [FAQs](https://smartconnections.app/connect-pro/faq/)
## Release notes
- Review AI-proposed changes in Obsidian - the new Connect Pro Inbox adds a human review step before supported delegated changes touch source notes.
- Correct exact claims with evidence - let an agent prepare a targeted replacement alongside the source evidence that motivated it.
- Move and rewrite excerpts - relocate exact text between notes or sections, with the option to refine what gets inserted at the destination.
- Reorganize complete sections - move blocks with their child sections in the exact order you approve, without silently rewriting their content or heading levels.
- Edit the proposal before applying it - refine wording, change destinations, reorder blocks, and inspect Changes / Before / After views from the same review.
- Protection against stale or unsafe applies - current source is rechecked before writing, stale plans require refresh, and unresolved recovery blocks unsafe retries.
- Agents can learn from prior reviews - proposal feedback exposes relevant applied, edited, and rejected history as evidence for better future proposals.
- Local MCP works independently of Remote GPT - use supported desktop MCP clients while the intended Obsidian vault is running, without requiring the legacy remote route.
- One Smart Tool surface across MCP and CLI - use the same configured capabilities from compatible desktop-agent and command-line workflows, with per-connector availability controls.
- A faster, cleaner review queue - ribbon access, pending-work indication, deduplication, History, Group, Label, and priority make delegated work easier to find and manage.
- Stronger failure and recovery handling - verified writes and conservative recovery make partial failures less likely to become destructive repeat attempts.
---
## Connect Pro
canonical: https://smartconnections.app/docs/connect-pro/
html_url: https://smartconnections.app/docs/connect-pro/
markdown_url: https://smartconnections.app/docs/connect-pro.md
llms_url: https://smartconnections.app/docs/connect-pro/llms.txt
last_modified: 2026-09-08T23:43:45.484Z
usage_notes: |-
Use this page to answer questions about Connect Pro.
excerpt: |-
Connect Pro Connect Pro 2.0 lets a connected AI client prepare corrections and reorganizations for review in Obsidian. Inspect the exact change, adjust it when needed, then apply or reject it in the Connect Pro Inbox. The dedicated proposal Tools do not change source notes when the client calls them. Connect Pro exposes configured Smart Tools from the running Obsidian Desktop vault through Local…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
- title: Graph
url: https://smartconnections.app/docs/graph/
# Connect Pro
Connect Pro 2.0 lets a connected AI client prepare corrections and reorganizations for review in Obsidian. Inspect the exact change, adjust it when needed, then apply or reject it in the **Connect Pro Inbox**. The dedicated proposal Tools do not change source notes when the client calls them.
Connect Pro exposes configured Smart Tools from the running Obsidian Desktop vault through Local MCP and Smart CLI Pro. Smart CLI is the short name for this CLI capability. The legacy Remote GPT connection is a separate route, not a prerequisite for Local MCP.
Connect Pro is not Sync, a mobile vault, unrestricted shell access, or an autonomous background agent. Inbox review applies to the dedicated proposal Tools, not every possible vault write.
## In this guide
- [Select and check a connection](#connect-pro-session)
- [Run a bounded action](#connect-pro-run-action)
- [Review proposed changes in the Inbox](#connect-pro-inbox)
- [Use proposal feedback](#connect-pro-feedback)
- [Understand approval, recovery, and stopping](#connect-pro-approvals)
- [Access from another device](#connect-pro-remote-device)
- [Use Smart CLI Pro](#smart-cli-pro)
## Select and check a connection
Keep the intended vault open in Obsidian Desktop with Connect Pro 2.0 enabled and its required Pro access active. The plugin requires Obsidian 1.12.2 or later. An **Active** plugin row means the plugin loaded, not that a connector or client is ready.
### Local MCP
In Connect Pro settings, enable **Local MCP server**. Make sure that it is **Running**, then open **Set up desktop clients** and use the endpoint and instructions shown for this installation. Connect only a client you trust. Refresh its available Tools and read one recognizable note in the intended vault before proposing changes.
The server and desktop client run on the same computer. Local MCP does not require **Remote GPT** to be **Connected**. Local transport does not guarantee local model processing: the client may send Tool inputs and returned content to its model provider.
*Local MCP is **Running** while Remote GPT is **Disconnected**. Use the endpoint shown in your own installation; a running listener does not establish client discovery or a successful vault read.*
### Smart CLI
For command-line and native CLI operations, enable Obsidian CLI and inspect **Advanced** > **Obsidian CLI integration** in Connect Pro settings. Smart CLI registers available configured Smart Tools with the host CLI integration. It does not expose unrestricted shell execution.
### Existing Remote GPT connection
When this route is available in the installed build and calling client, use the separate **Remote GPT connection** control. A **Connected** status and **Disconnect** control establish that remote session, not a successful workflow or the correct target vault. Read a recognizable note before writing.
The Official Connect Pro GPT is the legacy ChatGPT surface for this route. Its availability is separate from Local MCP readiness. A local **Running** indicator does not establish public MCP or cross-device access.
## Run a bounded action
Use the Tool list and schemas exposed by the current client. Select one exact note, block, folder, or named Context. Read current content and use discovered source or block keys; do not construct a block key from a heading title or assume a Tool exists because a task can be described in chat.
### Check Tool permissions
Open **Tool permissions** > **Tool actions** in Connect Pro settings. **Obsidian CLI** controls a listed action's availability through Obsidian CLI and Remote GPT. **Local MCP** independently controls its availability to local clients. New actions default to enabled, so review the available Tools rather than assuming a read-only starting configuration.
**Individual Obsidian CLI tools** controls which native CLI commands appear as individual Local MCP Tools. Those controls do not restrict the broad **Obsidian CLI** action, `obsidian_cli`. Disable that broad action as well when individual command restrictions are part of your intended boundary. These settings do not control unrelated registered Obsidian CLI commands.
An enabled control is not proof of client discovery or successful invocation. After changing availability, refresh client discovery. A disabled Smart CLI action is blocked immediately even if the command remains listed until Obsidian restarts.
*Each proposal Tool has separate **Obsidian CLI** and **Local MCP** controls. Enabled rows show availability, not invocation or a complete read-only boundary. The broad `obsidian_cli` action is outside this crop.*
### Distinguish the effect
Read-only Tools return information. Direct-write Tools can change the vault when invoked. The three proposal Tools below save pending review items instead; a successful proposal result is not an applied source change.
When a result and the vault state disagree, inspect the specific request before retrying. **Review and activity** > **Tool action requests** > **View requests** shows retained completed request payloads for the current Obsidian session. This request view is not the saved Inbox History or an approval queue.
## Review proposed changes in the Inbox
### Select the right proposal Tool
| Tool | Reviewable operation | Important boundary |
| --- | --- | --- |
| `propose_correction` | Replace one exact claim using exact reference evidence. | One independent phrase, sentence, list item, or table cell per call. The target excerpt must occur once within the supplied target. |
| `propose_move_excerpt` | Remove an exact excerpt and insert it at another target, optionally with revised wording. | This is a move, not a copy. Rewriting changes only the proposed insertion. |
| `propose_move_blocks` | Move complete selected blocks, including child sections, in the supplied order. | Blocks move without synthesis or heading rewriting. Whole-note keys, duplicate blocks, and overlapping parent/child selections are not valid inputs. |
Correction evidence must exist in the current reference source. Its presence does not prove that the claim is correct; inspect whether it actually supports the replacement. Use a separate proposal for each independent correction, and do not rewrite historical records merely because a later decision differs.
Moves explicitly select `destination_mode: append` for an existing note or block, or `destination_mode: create` for a new vault-relative `.md` or `.txt` note. A missing append target does not become a new note automatically. Missing destination sections are not created. A new destination note is created only when its proposal is applied.
### Recognize a pending result
The proposal call saves review data without editing source files. Its result includes the returned item `key`, `status: pending`, and `source_files_changed: false`.
*The native response reports `status: pending` and `source_files_changed: false`. In this historical capture, the first attempt timed out; this result followed an exact retry and **Allow once**. It is not an applied correction or an independent inspection of the source files.*
An identical still-valid pending proposal returns the existing key with `deduplicated: true`. This is exact pending deduplication, not semantic duplicate detection. Changing only the reason, priority, group, or label does not silently update the existing item. Rejection does not permanently suppress future proposals.
### Open and organize the Inbox
Open the Inbox from the ribbon action **Open Connect Pro Inbox**, the Command Palette action **Open: Connect Pro Inbox view**, or Connect Pro settings > **Review and activity** > **Connect Pro Inbox**. The settings button is **Review** when work is pending and **Open inbox** otherwise. The ribbon indicates pending work.
**Pending** contains proposals awaiting a decision. **History** contains applied and rejected outcomes. Pending items are ordered by priority, then oldest first; History shows the most recently reviewed items first. **Group** and **Label** filter independent items. Grouping never combines their approvals, and priority is review order, not confidence or authorization.
*One correction, one block move, and one rewritten excerpt move are **Pending**; **History** is empty. Group, Label, and priority organize independent reviews, not a batch approval.*
### Inspect and refine the result
Corrections and excerpt moves provide **Changes**, **Before**, and **After** views. A new destination has no **Before** view because the file does not exist yet. Inspect the supporting evidence, source removal, destination insertion, and affected files before applying anything.
Edit only the permitted replacement or insertion in the review. **Done editing** retains a local review draft without changing source files. **Cancel** restores the text from the start of that editing session. These draft edits are not saved as a final decision until a successful application.
***Done editing** keeps the proposed replacement as a review draft; it does not apply the correction. Check the evidence and refreshed diff before using the separate **Apply correction** action.*
For moves, **Change** opens the destination chooser. Select an existing note and location, or use **Create new note...**. You do not need to construct block-key syntax.
*After choosing an existing note, select an insertion location. This chooser does not itself apply the move, create a note, or change heading levels.*
For block moves, reorder the originally selected blocks; adding or removing blocks requires a new proposal.
Block moves show the selected content, included children, resulting heading placement, and exact file changes. They preserve the selected content and heading levels, with any joining separators shown in the preview. The source notes and unselected text remain. A moved heading does not automatically become a subsection of the destination, and links are not repaired.
A rewritten excerpt move requires separate approval of the original removal and the rewritten insertion. A relevant edit invalidates the preview and resets these confirmations. Make sure that the refreshed result still matches your intended change.
*Both confirmations are unchecked, so **Move and rewrite** is disabled. Review the rewritten insertion and exact original removal separately. **Link impact not checked** remains an explicit limit.*
### Apply, reject, or skip
Apply the current reviewed change with **Apply correction**, **Move excerpt**, **Move and rewrite**, or the block-count move button, as displayed. Apply rechecks the complete reviewed plan and current source snapshots, including correction reference evidence, before writing. Inspect the actual affected files after application.
**Reject** records a decision without changing source files. **Skip** leaves the proposal pending so you can return later. Applied and rejected outcomes remain in **History**.
*This captured session ends with two **Applied** moves and one **Rejected** correction. History records those decisions; inspect the actual notes for their resulting contents. It is not Undo.*
### Example: combine selected sections
This separate captured example moves two unchanged sections from **Reader interviews** and **Editorial plan** into the existing **Launch playbook**. It is not the block proposal in the mixed Inbox above. It demonstrates selected-block movement, not whole-note merging or synthesis.
The initial preview places **Reader need** before **Weekly format**. Check the destination, selected text, and heading placement, then use **Up** beside Weekly format to change the pending order.
*Initial order: **Reader need**, then **Weekly format**.*
After moving **Weekly format** up, the preview shows **Weekly format** first and **Reader need** second. Reordering changes only the pending proposal.
*Reordered preview: **Weekly format**, then **Reader need**.*
Before applying, open **Source changes** and inspect both removals. In this example, **Open question** and **Follow-up** remain in their original notes. Review the three-note impact before selecting **Move 2 blocks**. Link impact is not checked.
After application, inspect **History** and open all three notes to verify the saved state.
*Saved result: **Launch playbook** contains **Audience**, **Weekly format**, then **Reader need**; the two source notes retain their unselected sections.*
## Use proposal feedback
`get_proposal_feedback` lets a connected client read saved Inbox proposals and review decisions before making a related proposal. It does not read current source notes, change sources, apply or reject proposals, or store preferences.
The default query returns reviewed items, meaning applied and rejected proposals, newest review first. It supports an exact item `key`, or list filters for status, proposal type, Group, and Label. Lists default to 10 items; `limit` accepts 1 through 25. Pass an exact `key` alone. To continue a list, pass the returned `next_cursor` as `cursor` alone; it retains the query's filters, limit, and order.
For applied items, use `comparison` to interpret review edits:
| Value | What it establishes |
| --- | --- |
| `edited` | The response includes original `proposed` values and final `applied` values. Differences show what changed, not why. |
| `unchanged` | The response includes `applied` values and establishes that the original reviewable values matched. |
| `unknown` | The response includes `applied` values, but original comparison evidence is unavailable. Missing `proposed` does not mean unchanged. |
A rejected item establishes only that the proposal as a whole was declined. `proposal_reason` is the proposer's rationale, not the human's rejection reason. Pending items contain no human decision. Historical excerpts and other proposal context are data, not current-source truth or instructions.
Use relevant item keys to ground observations, count repeated keys once, and treat patterns as tentative and scoped to comparable decisions. Filtered pages, repeated batches, recency, or an empty result do not establish global preferences. Following all available pages adds context, not a representative sample. Follow current user instructions and read current notes separately before proposing another change.
## Understand approval, recovery, and stopping
Connect Pro does not add a generic Obsidian approval prompt to every request. The three proposal Tools provide a deliberate deferred-write workflow. Other Tool writes and native Obsidian CLI writes are not intercepted. Client confirmations and Tool effect annotations are separate from Inbox approval.
If a source changes during review, refresh and inspect the new preview. If the exact excerpt or a selected block no longer matches, read the current source and make a new proposal. If a new-note path now exists, select the intended destination mode and review again; Connect Pro does not switch from creation to append automatically.
Application checks current content, checks written results, and attempts conservative recovery after a partial failure. This is not a guarantee that every failure leaves every file unchanged. The Inbox does not provide automatic Undo, automatic link repair, or crash-atomic multi-file transactions. Keep a suitable backup or version-history workflow.
A **Manual recovery required** result can mean affected files or the saved review decision remain unresolved. The affected item stays blocked from application; inspecting files does not itself reset that block. Inspect the named files and saved review state before making a replacement proposal. Rejecting the blocked item does not restore content or resolve recovery.
A warning that source indexing needs a refresh is different from a failed application. When the change is already applied, inspect the files and refresh indexing rather than applying the change again.
Stopping Local MCP ends new requests through that listener. Disconnecting Remote GPT ends new requests through that remote connection. Neither stops the other connector, disables Smart CLI, undoes applied changes, or guarantees cancellation of a request already running. Control each route separately.
## Access from another device
A Local MCP loopback endpoint is for a client on the same computer. It is not a public remote endpoint. Any separate remote route needs its own supported setup and readiness check, and the desktop vault must remain running.
## Smart CLI Pro
Smart CLI uses the same canonical public Smart Tool contracts as MCP, including the proposal and feedback actions. The connector's command names can differ from Tool names. Inspect **Tool actions**, **Advanced** > **Obsidian CLI integration**, the available **Command examples**, and registered command metadata instead of guessing.
The v2 connector derives these default command names when the corresponding Tool is available and enabled:
| Smart Tool | Smart CLI command |
| --- | --- |
| `propose_correction` | `smart:propose:correction` |
| `propose_move_excerpt` | `smart:propose:move:excerpt` |
| `propose_move_blocks` | `smart:propose:move:blocks` |
| `get_proposal_feedback` | `smart:get:proposal:feedback` |
These names do not guarantee registration in every session. Use the installed command's input schema and metadata. The proposal result is still pending review; using CLI does not bypass the Inbox for a proposal action.
`smart:help` and `smart:skills` are not implemented in the supplied v2 source. Do not rely on older instructions that present them as available commands.
## Troubleshooting
| What you see | First check |
| --- | --- |
| Local MCP is enabled but not Running | Inspect the displayed error, then turn **Local MCP server** off and on to retry. |
| Remote GPT is Disconnected but Local MCP works | These are independent connectors; this is not by itself a failure. |
| A Tool is missing | Inspect required plugins, channel permissions, and host CLI availability when needed, then refresh client discovery. |
| A proposal succeeded but notes did not change | Open the Inbox; pending review is the expected result. |
| A correction excerpt matches more than once | Read the target and supply a more specific exact excerpt or discovered block key. |
| Apply is unavailable | Inspect preview errors, required confirmations, source changes, or recovery feedback. |
| A proposed block changed | Read the current block and make a new proposal; do not approve the old selection. |
| A change is applied but retrieval is stale | Inspect the file and refresh source indexing; do not reapply the change. |
| Older contradiction, Extract, or Merge items are absent | The unified Inbox loads its native store and does not import those separate experimental stores. |
## Related documentation
- [Getting started with Connect Pro](https://smartconnections.app/connect-pro/getting-started/)
- [Connect Pro FAQs](https://smartconnections.app/connect-pro/faq/)
- [Smart Plugins documentation](https://smartconnections.app/docs/plugins/)
---
## Getting Started
canonical: https://smartconnections.app/connect-pro/getting-started/
html_url: https://smartconnections.app/connect-pro/getting-started/
markdown_url: https://smartconnections.app/connect-pro/getting-started.md
llms_url: https://smartconnections.app/connect-pro/getting-started/llms.txt
last_modified: 2026-09-08T23:43:45.480Z
usage_notes: |-
Use this page to answer questions about Getting Started.
excerpt: |-
Getting started with Connect Pro Start with one read that matches the intended vault. Then use a disposable target to complete the Connect Pro 2.0 propose-review-apply workflow. A pending proposal is the first result, not a completed source change. 1. Open the intended desktop vault Use Obsidian Desktop 1.12.2 or later with Connect Pro 2.0 enabled and its required Pro access active. Keep the…
suggested_links:
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
# Getting started with Connect Pro
Start with one read that matches the intended vault. Then use a disposable target to complete the Connect Pro 2.0 propose-review-apply workflow. A pending proposal is the first result, not a completed source change.
## 1. Open the intended desktop vault
Use Obsidian Desktop 1.12.2 or later with Connect Pro 2.0 enabled and its required Pro access active. Keep the intended vault open. An **Active** plugin row confirms loading, not client connectivity.
## 2. Select one calling route
For Local MCP, open Connect Pro settings and enable **Local MCP server**. Make sure that it is **Running**, then use **Set up desktop clients** with the endpoint shown for this installation. Follow the installed instructions for your supported client on the same computer. Remote GPT does not need to be connected.
*Local MCP is **Running** while Remote GPT is **Disconnected**. Use the endpoint shown in your own installation; a running listener does not establish client discovery or a successful vault read.*
For Smart CLI Pro or native CLI operations, enable Obsidian CLI and inspect **Advanced** > **Obsidian CLI integration**. Smart CLI uses the same underlying Smart Tool contracts. An existing Remote GPT integration uses its own connection and support requirements; do not mix its readiness checks with Local MCP.
Open **Tool permissions** > **Tool actions** and review availability for your selected channel. New actions default to enabled. For a proposal-only write workflow, retain the reads you need and the selected proposal action, and disable unrelated direct-write actions exposed through that channel. In Local MCP, also disable the broad `obsidian_cli` Tool when you need individual CLI restrictions; its access is not restricted by the individual command toggles.
Only connect a client you trust. Local MCP describes local transport, not a guarantee that the client's model provider receives no vault content.
## 3. Check one read-only result
Refresh the client's available Tools. Read one recognizable note by its exact discovered key and compare the returned text with that note in Obsidian. Do not expand scope or write when the route or target is uncertain.
## 4. Propose one correction
Select a disposable target note and a reference note that contains a clearly different, accurate claim. Make sure that the target and reference excerpts are current and exact. Use one small phrase or sentence, not a whole-page rewrite.
For example, manually prepare two disposable notes: a reference stating `The review happens on Friday.` and a target stating `The review happens on Monday.` Then use this request, replacing the bracketed names with those notes:
> Read [reference note] and [target note]. Use propose_correction to propose replacing the target's Monday claim with the exact Friday claim supported by the reference. Use one correction, Group "Connect Pro first review", and Label "Setup test". Do not use a direct-write Tool or native CLI write. Leave the change pending for me to review in Obsidian and report the returned item key.
The Tool result should include `status: pending` and `source_files_changed: false`. Open the target note and make sure that its text is still unchanged. Do not repeat the request merely because the note has not changed yet.
For a first move instead, use `propose_move_excerpt` for one exact passage or `propose_move_blocks` for complete discovered blocks. Select `append` or `create` explicitly. A new destination note should not exist until application.
## 5. Review in Obsidian
Open the ribbon action **Open Connect Pro Inbox**, the Command Palette action **Open: Connect Pro Inbox view**, or Connect Pro settings > **Review and activity** > **Connect Pro Inbox** > **Review**.
Inspect the target, rationale, reference evidence, and **Changes**, **Before**, and **After** views. A reference excerpt is supporting evidence to evaluate, not a guarantee that the proposed claim is correct.
Edit the proposed replacement in place when needed. **Done editing** retains the local draft without applying it; **Cancel** restores the text from the start of that editing session. Wait for a current preview before deciding.
For moves, **Change** selects an existing note and location or **Create new note...**. A new destination has no **Before** view because the file does not exist yet. Rewritten excerpt moves need separate removal and rewritten-insertion confirmations. For block moves, inspect included children, final order, heading placement, and exact file changes. A moved heading is not automatically nested or rewritten.
## 6. Apply or reject and inspect the outcome
For the correction, select **Apply correction** only when the current preview is ready. Compare the actual target note with the reviewed result and make sure that the reference note is unchanged. The item should appear as **Applied** in **History**.
**Reject** records a decision without editing source files. **Skip** leaves the proposal pending. For this first test, an applied change that matches the preview or an intentionally rejected proposal with unchanged sources is a complete reviewed result.
When a preview is stale, inspect a fresh preview. When an exact excerpt or selected block no longer matches, read current content and make a new proposal. A manual-recovery warning requires inspection of the named files and saved review state; it is not permission for a blind retry. The blocked item does not become applicable merely because you inspected it.
The Inbox has no automatic Undo or automatic link repair. Recovery protection is not a crash-atomic multi-file transaction. These limits apply even to a small test.
## 7. Read the decision and stop the route you used
When `get_proposal_feedback` is available and enabled, ask the client to read the returned item key. It should describe the recorded decision. For an applied item, inspect `comparison` and the final `applied` values. A rejected item does not contain a human rejection reason merely because it contains `proposal_reason`.
Stop Local MCP when finished with the local client, or disconnect Remote GPT when finished with that route. They are independent controls, not a global cancellation or Undo mechanism. They do not disable Smart CLI. Applied changes remain, and an already-running request is not guaranteed to be canceled.
## Example: review a rewritten excerpt move
The Monday/Friday correction above is a disposable exercise, not a recorded run. This separate historical example follows one proposed move from **Launch working notes** into a new **Reader voice** note. It illustrates the review checkpoints, not completion of the exercise or a new v2 acceptance run.
### Choose the suggestion
In **Pending**, select **Review** beside **Move excerpt to Reader voice**. Other pending proposals have their own decisions.
*Follow the **Reader voice** proposal. The correction and block move have their own independent decisions.*
### Adjust the insertion
Check the destination. Use **Edit insertion** to adjust the proposed text, then **Done editing**. Editing this draft does not create Reader voice or remove the source excerpt.
***Done editing** keeps a local draft. It does not create Reader voice or remove the original excerpt; a new destination has **Changes** and **After**, but no **Before** view.*
### Review both changes
Read the final insertion and the exact original excerpt marked for removal. This is a move, not a copy. Reject the suggestion when the proposed changes are not useful.
*Both confirmations are unchecked, so **Move and rewrite** is disabled. Review the rewritten insertion and exact original removal separately. **Link impact not checked** remains an explicit limit.*
### Confirm and apply
For this rewritten move, check **Approve inserting the rewritten text** and **Approve removing the original excerpt**. Select **Move and rewrite** only after both changes match your intent and the preview is ready.
*Both confirmations are checked and the action is ready. This is still a pending proposal, not an applied move or an already-created destination.*
### Verify the outcome
In **History**, open **Details** for the Applied Reader voice move. Compare **Original excerpt** with **Insertion**, then open the source and destination notes to verify removal and insertion. The stored original excerpt is historical text, not proof that it remains in the source.
***Applied** records the decision and final insertion. **Original excerpt** is retained historical text; inspect both notes to verify the current removal and insertion.*
These screenshots are checkpoints from one review session, not a continuous recording, so Pending and History counts change between steps. Follow the **Reader voice** proposal and note names rather than the counts.
## Continue after the first result
Use a named project or small review batch. **Group** and **Label** help locate independent decisions; priority affects their order, not confidence or shared approval. Expand only after the read, pending proposal, human decision, and resulting file state consistently match.
Use prior feedback as scoped historical evidence. Read current notes and follow current instructions for each new proposal. The history is not automatic preference learning or authorization to apply changes.
See the [documentation](https://smartconnections.app/docs/connect-pro/) for Tool contracts and boundaries and the [FAQ](https://smartconnections.app/connect-pro/faq/) for pending results, duplicates, recovery, and connection distinctions.
---
## Faq
canonical: https://smartconnections.app/connect-pro/faq/
html_url: https://smartconnections.app/connect-pro/faq/
markdown_url: https://smartconnections.app/connect-pro/faq.md
llms_url: https://smartconnections.app/connect-pro/faq/llms.txt
last_modified: 2026-09-08T23:43:45.476Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Connect Pro FAQs What is Connect Pro?#Connect Pro exposes configured Smart Tools from a running Obsidian Desktop vault through Local MCP and Smart CLI Pro. In 2.0, its unified Connect Pro Inbox lets you review AI-proposed corrections and moves before those proposals change source notes. The legacy Remote GPT connection is separately controlled. Connect Pro is not Sync, a mobile vault,…
suggested_links:
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
# Connect Pro FAQs
## What is Connect Pro?
Connect Pro exposes configured Smart Tools from a running Obsidian Desktop vault through Local MCP and Smart CLI Pro. In 2.0, its unified **Connect Pro Inbox** lets you review AI-proposed corrections and moves before those proposals change source notes. The legacy Remote GPT connection is separately controlled.
Connect Pro is not Sync, a mobile vault, unrestricted shell access, or an autonomous agent. Direct-write Tools outside the dedicated proposal workflow are not automatically held for review.
## What changes in Connect Pro 2.0?
The central workflow is propose, review, then apply or reject. Three Tools cover exact corrections, excerpt moves with optional rewritten insertion, and complete block moves. The Inbox adds editable previews, destination and block-order controls, Pending and History views, Group and Label filters, and priority-based review order.
`get_proposal_feedback` exposes saved proposals and decisions as historical evidence. Local MCP and Smart CLI use the same Smart Tool contracts, with separate connector availability controls. Source checks, checked writes, and conservative recovery protect the application path without promising Undo or crash-atomic transactions.
## Does Connect Pro require anything extra?
Keep the intended vault open in Obsidian Desktop 1.12.2 or later with Connect Pro 2.0 enabled and its required Pro access active. For Local MCP, start **Local MCP server** and connect a trusted supported client on the same computer using **Set up desktop clients**.
Enable Obsidian CLI for Smart CLI and native CLI operations; it is not a blanket prerequisite for every Local MCP Smart Tool. An **Active** plugin row does not prove that a connector or client is ready. Read one recognizable note before proposing a change.
## Does Local MCP require Remote GPT to be connected?
No. Local MCP has its own **Running** state and endpoint. **Remote GPT** can remain **Disconnected** while a local client uses enabled Tools. Disconnecting Remote GPT does not stop Local MCP, and stopping Local MCP does not disconnect Remote GPT or disable Smart CLI.
A running listener is not proof of client discovery or a successful read from the intended vault. Check those separately.
*Local MCP is **Running** while Remote GPT is **Disconnected**. Use the endpoint shown in your own installation; a running listener does not establish client discovery or a successful vault read.*
## What actions can Connect Pro run?
Use the current client's Tool list and Connect Pro settings > **Tool permissions** > **Tool actions**. Installed capabilities can supply retrieval, exact reads, Context workflows, native CLI operations, direct writes, and explicit reviewable proposals. Availability depends on installed plugins, the current host session, and per-channel controls.
The proposal Tools are `propose_correction`, `propose_move_excerpt`, and `propose_move_blocks`. The read-only feedback Tool is `get_proposal_feedback`.
## Do the individual CLI controls restrict every CLI route?
No. **Obsidian CLI** and **Local MCP** are separate availability controls for listed Smart Tools. **Individual Obsidian CLI tools** controls the native commands exposed as individual Local MCP Tools.
The broad **Obsidian CLI** action, `obsidian_cli`, is not restricted by those individual command settings. Disable it as well when you need that narrower boundary. These controls do not restrict unrelated registered Obsidian CLI commands, and new Smart Tool actions default to enabled.
An enabled control does not prove discovery or execution. A disabled Smart CLI action is blocked immediately, although it can remain listed until Obsidian restarts.
*Each proposal Tool has separate **Obsidian CLI** and **Local MCP** controls. Enabled rows show availability, not invocation or a complete read-only boundary. The broad `obsidian_cli` action is outside this crop.*
## Does Connect Pro ask before every action?
No. Client confirmations belong to the client, and Tool effect annotations describe effects rather than create an approval boundary. Unrelated direct-write and native CLI actions are not held in the Inbox.
`propose_correction`, `propose_move_excerpt`, and `propose_move_blocks` save pending proposals without changing source files. Their source changes happen only when you apply the reviewed proposal in Obsidian.
## Why did a proposal succeed without changing my note?
That is expected. The result includes `status: pending`, `source_files_changed: false`, and the saved item `key`. Open the ribbon action **Open Connect Pro Inbox**, the Command Palette action **Open: Connect Pro Inbox view**, or settings > **Review and activity** > **Connect Pro Inbox**.
A new destination note is created only when the proposal is applied. Pending review is a successful proposal outcome, not a missing write to retry.
## Which proposal Tool should I use?
Use `propose_correction` for one exact claim supported by an exact reference excerpt. Use one independent phrase, sentence, list item, or table cell per call. The target excerpt must occur exactly once within the selected target. Supporting evidence still needs human evaluation.
Use `propose_move_excerpt` for one exact passage that should move, optionally with different insertion wording. Use `propose_move_blocks` for complete discovered blocks and their children, in the desired final order. Block moves do not synthesize content, rewrite headings, delete source notes, or repair links.
Use returned source and block keys rather than constructing them. Overlapping parent and child selections are not valid block-move inputs.
## Can I correct a proposal before applying it?
Yes. Corrections and excerpt moves offer **Changes**, **Before**, and **After** views. A new destination has no **Before** view because the file does not exist yet. Edit the permitted replacement or insertion in the review, select another destination with **Change**, or reorder the originally selected blocks.
**Done editing** retains a local draft without applying or saving a final decision. **Cancel** restores the text from the start of the editing session. Relevant edits require a fresh preview. Adding or removing selected blocks requires a new proposal.
***Done editing** keeps the proposed replacement as a review draft; it does not apply the correction. Check the evidence and refreshed diff before using the separate **Apply correction** action.*
## Why does a move need two confirmations?
An excerpt that is both moved and rewritten requires separate approval of the original removal and the rewritten insertion. Both must be satisfied before **Move and rewrite** is available. A relevant edit resets them.
An unchanged excerpt move uses **Move excerpt**. Plain block moves preserve selected text and heading levels; they do not silently turn moved headings into subsections of the destination.
*Both confirmations are unchecked, so **Move and rewrite** is disabled. Review the rewritten insertion and exact original removal separately. **Link impact not checked** remains an explicit limit.*
## Can a move create a new note?
Yes, with an explicit new-note destination. The Tool uses `destination_mode: create` with a new vault-relative `.md` or `.txt` path. In the Inbox, use **Change** > **Create new note...**. Creation happens only on application.
`destination_mode: append` requires an existing note or block. Missing append targets and missing sections are not created automatically. When a new-note path becomes occupied during review, select the intended mode and destination and inspect a fresh preview.
## Does repeating a proposal create duplicates?
An identical still-valid pending proposal returns the existing key with `deduplicated: true`. This is exact pending deduplication, not semantic similarity detection. Changing only the reason, priority, group, or label does not silently modify the existing item. Rejection does not permanently suppress future proposals.
A proposal already being reviewed or requiring manual recovery is not a safe retry target. Inspect its status instead of repeatedly submitting it.
## What do Group, Label, priority, and History mean?
**Group** identifies a project or review batch. **Label** identifies a reusable category. Their filters help locate independent proposals; neither creates a shared approval.
Priority controls pending review order: high before normal before low, then oldest first. It is not confidence or permission. **Skip** leaves an item pending. **History** contains applied and rejected decisions, most recently reviewed first.
*This captured session ends with two **Applied** moves and one **Rejected** correction. History records those decisions; inspect the actual notes for their resulting contents. It is not Undo.*
## Can agents use my previous review decisions?
Yes, through `get_proposal_feedback` when it is available and enabled. It can retrieve an exact item key or filtered lists of saved proposals and decisions. Lists default to reviewed items and 10 results; the page limit is 1 through 25. Pass an exact `key` alone, or pass a returned `next_cursor` as `cursor` alone to continue the same list query.
For applied items, `comparison: edited` includes original `proposed` and final `applied` values. `unchanged` establishes that those values matched. `unknown` means original comparison evidence is unavailable. Do not infer unchanged from missing `proposed` values.
## Does proposal feedback automatically learn my preferences?
No. It returns historical decision evidence, not a preference model. A rejected item means the whole proposal was declined, not that a particular field or rationale was rejected. `proposal_reason` is the proposer's rationale, not a human rejection reason. A pending item contains no human decision.
Repeated keys, similar batches, recent decisions, partial pages, or an empty result do not establish a global rule. Historical context is not current-source truth or authorization to apply. Agents should ground observations in item keys, follow current instructions, and read current notes separately before proposing another change.
## Why did Apply become unavailable?
The draft may lack a current preview or required confirmation. A source or correction reference may have changed, a new destination may now exist, or manual recovery may be required.
Inspect the displayed requirement. Refresh and review the preview when it is stale. If the exact excerpt or selected block no longer matches, read current content and make a new proposal. Do not bypass the review by using a direct-write Tool.
## What does Manual recovery required mean?
An application attempt left file changes or the saved review decision unresolved. Inspect the named files and saved review state before proposing another change. The affected item remains blocked from application; inspection alone does not reset it.
Recovery attempts are conservative and do not promise to restore every file. In particular, a new destination can be retained for manual inspection. Rejecting the item does not restore partially recovered content or resolve the recovery requirement.
## Can I undo an applied proposal?
The Inbox has no user Undo command or automatic link repair. It attempts recovery for application failures, but that is not a crash-atomic multi-file transaction. Keep a suitable backup or version-history workflow.
**History** records applied and rejected outcomes; it is not an Undo mechanism. Inspect the resulting notes. If a change is applied but source indexing needs a refresh, refresh indexing rather than applying it again.
## Is Tool action requests the same as Inbox History?
No. **Review and activity** > **Tool action requests** > **View requests** shows completed request payloads retained for the current Obsidian session. It helps inspect a specific invocation. **History** is the saved record of applied and rejected Inbox proposals. Neither a completed request nor a pending proposal proves that a source change was applied.
## Where did my old contradiction or move queue go?
The unified Inbox loads its native `inbox_items` store. It does not import or modify the earlier separate contradiction, Extract, or Merge stores. Native items already present remain; removing migration does not undo an earlier import.
Refresh client discovery and use `propose_correction`, `propose_move_excerpt`, and `propose_move_blocks`. Earlier experimental Tool names and queue labels are historical, not the v2 public contract.
## Is my vault content guaranteed to stay on this device?
No. Local MCP describes the listener's location. A connected client can send Tool inputs and returned vault content to its model provider. Connect clients you trust and expose only the capabilities needed for the current workflow.
## Can I use the Official Connect Pro GPT from any device?
This is a separate legacy route, not a Local MCP feature. It requires an available supported GPT integration, a connected Remote GPT session, and the running desktop vault. Local MCP readiness alone does not establish remote or cross-device support.
## Why does the Official Connect Pro GPT say my vault is unavailable?
For that route, inspect **Remote GPT** status, the intended running vault, account requirements, and native CLI availability. Do not use Remote GPT diagnostics to decide whether Local MCP is running. Stop before writing when the route or target is uncertain.
## Is Connect Pro the same as Smart Chat?
No. Smart Chat owns its in-Obsidian conversation experience. Connect Pro exposes bounded vault capabilities to an external calling client. The Inbox owns human review of the dedicated correction and move proposals, not every Smart Chat or vault operation.
## Does installing Connect Pro migrate an older Smart Connect or Custom GPT Actions setup?
No automatic connection migration is established by the v2 Inbox work. Make sure that a replacement route works before dismantling a working legacy connection. Connection migration and old Inbox-record migration are different concerns.
## Does this canon define a separate public API or OpenAPI workflow?
Local MCP is an implemented local connector, and Smart CLI is a registered command surface. Neither establishes a public hosted MCP endpoint or a separate public OpenAPI setup. A loopback listener is not publicly reachable merely because it is **Running**.
## Where is the Smart Connect app?
Connect Pro is the current plugin direction in this project. Older Smart Connect material is historical setup context. Use the installed instructions and check the replacement route rather than assuming compatibility.
## Where can I learn more?
- [Connect Pro documentation](https://smartconnections.app/docs/connect-pro/)
- [Getting started with Connect Pro](https://smartconnections.app/connect-pro/getting-started/)
- [Pro Plugins FAQ](https://smartconnections.app/pro/faq/)
---
## Connections
canonical: https://smartconnections.app/docs/connections/
html_url: https://smartconnections.app/docs/connections/
markdown_url: https://smartconnections.app/docs/connections.md
llms_url: https://smartconnections.app/docs/connections/llms.txt
last_modified: 2026-09-08T22:58:03.888Z
usage_notes: |-
Use this page to answer questions about Connections.
excerpt: |-
Smart Connections Find related notes from the current note, an exact block, the note footer, or an Obsidian Base. The Connections view and Footer connections are included in Core and remain available when Pro is active. Inline connections and Connections in Bases require Pro. New to Smart Connections? Start with Getting started with Smart Connections. Open the Connections view. Complete one…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
- title: Graph
url: https://smartconnections.app/docs/graph/
# Smart Connections
Find related notes from the current note, an exact block, the note footer, or an Obsidian Base. The Connections view and Footer connections are included in Core and remain available when Pro is active. Inline connections and Connections in Bases require Pro.
> [!TIP] New to Smart Connections?
> Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/). Open the Connections view. Complete one note-first discovery. Then use the detailed controls below.
## In this guide
- [Find related notes in Obsidian with Smart Connections (Core and Pro)](#smart-connections-view)
- [Find related notes beside the text you are writing (Pro)](#smart-connections-inline)
- [Show related notes at the bottom of an Obsidian note (Core and Pro)](#smart-connections-footer)
- [Tune Smart Connections scoring and ranking (Pro)](#smart-connections-scoring)
- [Rank Obsidian Bases rows by semantic relevance (Pro)](#smart-connections-bases)
## Find related notes in Obsidian with Smart Connections
**Edition: Core and Pro.**
The Connections view is note-anchored semantic discovery: while auto-refresh is running, the current note determines the related notes or blocks shown beside it.
*The ranked list is the primary workflow. The mini graph appears when **Version 4.0 (Graph + List)** is selected. This is a Core Connections display mode, not the separate Smart Graph product.*
What the numbers identify:
1. The note in the editor supplies the current comparison text.
2. The panel header names the Connections target.
3. The mini graph shows the current candidate neighborhood.
4. Ranked rows preserve readable source identities and relative scores.
*Use the populated list to choose a result to inspect. Scores are relative signals, and the mini-graph lines are not authored links.*
### Open the Connections view
Open the Command Palette. Run the command for your installed edition:
- Core: `Smart Connections: Open: Connections view`
- Pro: `Smart Connections Pro: Open: Connections view`
What the numbers identify:
1. The full Pro command in the Command Palette search field.
2. The matching **Open: Connections view** command ready to run.
Core uses the same command suffix under the **Smart Connections** prefix.
The Smart Connections ribbon action opens the same Connections view when that entry is available. Run the command or select the ribbon action. Wait for the populated Connections view.
The default view uses the active note as the reference source and updates when that source changes while auto-refresh is running.
#### When should I use Smart Lookup instead?
Use Lookup when a question or typed idea should determine the results. Use Connections when the current note is the anchor.
### Auto-refresh and fixed-target behavior
When the active-state menu offers **Pause auto-refresh**, active-note changes drive the result set. Select **Pause auto-refresh** when one reference note must remain fixed while you open or compare results.
If the list appears stuck after changing notes, examine the auto-refresh menu state. If auto-refresh is paused, use the available resume action.
What the numbers identify:
1. **Beta Reader Feedback** is open in the editor.
2. **Landing Page Messaging** remains the fixed Connections target.
3. The play control resumes auto-refresh.
4. The ranked result set remains attached to the fixed target.
*The target stays fixed while another note is open. Resume auto-refresh when you want results to follow the active note again.*
#### Why did the list change when I opened a result?
The Connections view follows the active note while auto-refresh is running. Select **Pause auto-refresh** before opening results when one reference note must remain fixed.
### Drop an item anywhere in Connections to change the target
Drop one indexed note or block anywhere inside the live Connections view. Valid drop areas include the top bar, target label, graph, result list, and background.
The target menu groups recently used notes under **History** and heading or line-range targets under **Blocks**. Changing the target does not change whether result rows use Sources or Blocks.
Accepted sources include a File Navigator note and a Connections, Lookup, or graph result. The view rerenders around the dropped target.
Folders, unindexed items, ambiguous paths, and multi-item drops are rejected because one Connections view has one target.
### Refresh connections
**Refresh connections** recomputes results for the current reference note.
Use it after substantial note edits, a settings change, or a stale result state.
What the numbers identify:
1. **Pause auto-refresh** keeps the current target fixed while you inspect another source.
2. **Copy as list of links** copies the visible result set.
3. **Open random connection** opens one candidate for review.
4. **Scoring algorithm** opens the Pro primary-score selector. Ranking and settings remain visible below it.
*Review the visible result set. Make sure that the destination is correct. Then choose an action.*
**Send to Smart Context** requires Smart Context. **Explore in Smart Graph** requires Smart Graph. Scoring and ranking controls require Connections Pro. Check the Store when a handoff is unavailable.
### Read result rows
Each row represents a related Smart Source or Smart Block.
A row can show:
- source or block label
- path or heading context
- relative connection score
- expand/collapse state
- pinned or hidden feedback state where applicable
What the numbers identify:
1. The panel target is **Landing Page Messaging**.
2. The mini graph visualizes the current candidate set.
3. The expanded row keeps its source identity and relative score.
4. The expanded Markdown exposes source text for review.
*The score is relative to this result set. Expand the row to inspect the source before deciding whether to use it.*
### Expand and collapse a result
Expand a result to inspect more source content without opening the note.
**Expand all results** and **Collapse all results** apply the same presentation state to the visible rows.
### Hover preview
On desktop, hold Cmd/Ctrl while hovering a result to use Obsidian Hover Preview.
Preview lets you confirm relevance without changing the active note or the list anchor.
### Open a result
Clicking a result follows Obsidian's default link behavior:
- Click: open in the current pane.
- Cmd/Ctrl + click: open in a new tab.
- Cmd/Ctrl + Alt + click: open in a new pane or split.
### Drag a result into a note or another Smart Plugin
Drag a result to:
- a Markdown editor to create an Obsidian link (Core and Pro)
- Smart Chat to add it to the current response context (requires Smart Chat)
- an existing named-context dashboard row to curate it into that saved context (requires Smart Context)
- Smart Graph to add its indexed source to the current graph scope (requires Smart Graph)
- the live Connections view to make it the single current target (Core and Pro)
Use the editor route when a suggested relationship should become an authored relationship in the current note.
*The preview supports the decision. The inserted Obsidian link makes the reviewed relationship durable.*
### Copy as a list of links
**Copy as list of links** copies the current result set as Obsidian links.
The output is suitable for a Related section, reading trail, project hub, or review note.
### Send to Smart Context
**Send to Smart Context** opens a reviewable context set from the current Connections results.
Remove noisy results before copying or using the set in an AI workflow.
This action requires Smart Context. If Context is unavailable:
1. Open the Smart Plugins Store.
2. Install Smart Context if it is not installed.
3. Enable Smart Context if it is disabled.
4. If the Store shows **Reload required**, complete that action.
5. Return to Connections.
> [!IMPORTANT] Verify the destination
> After sending results, open Smart Context. Make sure that the source identities are correct before you copy or send them onward.
What the numbers identify:
1. **Undo** reverses the handoff before you copy or send the reviewed set onward.
2. The summary reports 15 sources, the estimated token size, and one rule.
3. The expanded **Newsletter Launch** tree preserves the transferred source hierarchy and names.
4. **Copy context** is available after review.
*Sending results creates a reviewable source set. Confirm the identities before copying or sending it onward.*
### Hide a result
Hide removes a recurring low-value result from the visible list.
In Connections Pro, hidden feedback can also affect feedback-aware scoring algorithms when one is selected.
The root list menu exposes **Unhide All (n)**, where `n` is the current hidden-result count. Its enabled or disabled state shows whether a bulk reversal is available. After using it, refresh the same target and make sure that the intended results return.
*The menu reports the current hidden and pinned counts and enables **Unhide All** or **Unpin All** when those states exist.*
### Pin a result
Pin keeps an important result at the front of the list.
Pinned results remain at the front while still showing a score for the current reference note.
The root list menu exposes **Unpin All (n)**, where `n` is the current pinned-result count. Its enabled or disabled state shows whether a bulk reversal is available. After using it, refresh the same target and make sure that the intended pin state appears.
### Interpret connection scores
The score is a relative ranking signal for the current result list.
- Compare scores inside the same list.
- Do not treat the score as an absolute grade.
- Preview or open the source before using it.
- Score ranges vary by model, candidate pool, and algorithm.
#### Does a high score mean the notes should be linked?
No. It means the result ranked highly for the current comparison. Add a link only after the relationship helps the note.
### Choose whole-note Sources or smaller Blocks
**Sources** returns whole-note candidates. **Blocks** returns heading or block candidates.
Start with Sources for broad discovery. Use Blocks when long notes hide the useful section.
### Hide notes already linked from the current note
**Edition: Pro.**
Use **Exclude outlinks** when the current note already links to sources that should not be suggested again. This setting filters the displayed Connections list. It does not delete links, remove notes from the vault, or change Smart Environment indexing scope.
Workflow:
- Start from a note that links to one or more current Connections results.
- Open Connections Pro settings.
- Enable **Exclude outlinks**.
What the numbers identify:
1. The description limits **Exclude outlinks** to displayed Connections results. It does not change Smart Environment indexing.
2. The enabled toggle applies that display filter to the current anchor.
After enabling it:
1. Return to the same anchor.
2. Make sure that linked notes are absent.
3. Make sure that unrelated results remain.
Disable **Exclude outlinks** when you want linked notes to be eligible again.
See [the filter explanation](https://smartconnections.app/smart-connections/faq/#how-do-i-hide-notes-that-are-already-linked-from-the-current-note) for the control and comparison steps.
### Troubleshoot empty or stale results
Use the smallest recovery that matches the state.
| State | First check |
| --- | --- |
| Empty on a tiny note | Open a note with enough meaningful text. |
| Not updating | Confirm auto-refresh is running. |
| Stale after edits | Use **Refresh connections**. |
| Smart Environment not loaded | Select **Load Smart Environment**.
Wait for **Smart Environment ready** or **Ready**. |
| Embedding paused | Open the status view.
Select **Resume embedding**. |
| One expected note missing | Keep the note active.
Select **Inspect active note** from the status menu. |
| Several expected notes missing | Select **Show stats** from the status menu.
Alternatively, select **Environment stats** from the status view.
Check source eligibility.
Check the current embeddings. |
| Queued re-import work | Open the status view.
Select **Run re-import**. |
| Broad or noisy | Tune result type and limits first. Use Pro filters or scoring only after one useful result is confirmed. |
What the numbers identify:
1. Total indexed items.
2. Remaining and unexpected embedding work.
3. Smart Sources eligibility and current coverage.
4. Smart Blocks eligibility and current coverage.
*Coverage shows preparation status. Relevance still depends on the active note and Connections settings.*
#### Inspect one active note with Source Inspector
When one meaningful note is unexpectedly missing:
1. Keep that note active.
2. Choose **Inspect active note** from the Smart Environment status menu.
3. In Source Inspector, make sure that the note is eligible.
4. Make sure that its required Smart Source or Smart Block state is current.
5. Correct its inclusion or content before you change Connections Pro ranking controls.
#### Are similar results duplicates?
Not necessarily. Use Smart Dedupe when repeated material needs side-by-side review.
### Connections sidebar behavior
The sidebar owns active-file refresh, pause state, folding, settings access, and visibility recovery.
#### Auto-open behavior
New installations can open the Connections sidebar automatically so the first results are visible.
Existing workspaces retain normal Obsidian view persistence.
#### Active-file refresh
While auto-refresh is running, the sidebar refreshes when the active note changes.
The view also updates after **Refresh connections** and relevant Connections-list settings changes.
#### Fixed-target behavior
**Pause auto-refresh** keeps the current reference source while the active editor changes.
While auto-refresh is paused, active-note changes do not refresh the list.
##### Why are results still anchored to an older note?
Auto-refresh may be paused. Turn it back on from the same control to resume active-note updates.
#### Settings gear
The gear control opens the relevant Connections settings.
Settings changes that affect list filters trigger a result refresh.
#### Expand or collapse all results
The root list menu shows **Expand all results** or **Collapse all results**, depending on the current row state. These actions change the expanded state of visible result rows without changing the result set.
#### Sidebar visibility recovery
When the sidebar is hidden, it defers refresh work. When it becomes visible again, it updates if the active note changed and auto-refresh is running.
##### Why did the sidebar not update while it was hidden?
Hidden views defer unnecessary work. The pending source is checked when the view becomes visible again.
### Connections mini graph
**Version 4.0 (Graph + List)** is a Core Connections display mode. It adds the Connections mini graph above the ranked result list. The mini graph represents the current candidate set and is not the separate Smart Graph product.
The main Connections list and Footer can each use **List only** or **Version 4.0 (Graph + List)**. Graph visibility follows the selected component. There is no separate **Show graph** setting.
Use the mini graph to scan the result set. Then use the list to open a source. Review the source before you keep the relationship.
#### Connections graph
The graph displays the current Connections result set as nodes arranged around temporary semantic groups. Use it to examine the shape of the results. Then use the list to review a source. Act on the source only after this review.
#### Node states
Pinned results use the Connections accent treatment. Hidden results remain visible as muted nodes so prior feedback is still recognizable in the graph.
##### Why can a hidden result still appear?
The graph can keep it as a muted node so prior feedback remains visible.
#### Graph lines
Lines visualize relationships among the displayed results. They do not create Obsidian links and should be treated as exploratory signals.
##### Do graph lines create links in my notes?
No. They are visual signals. Add a normal Obsidian link when the relationship should persist.
#### Preview a node
On desktop, hold Cmd/Ctrl while hovering a node to use Obsidian Hover Preview for the corresponding note or block.
#### Select a node
Selecting a node scrolls the matching Connections row into view and expands it for review.
#### Semantic groups
Group placement is recalculated from the current result set. A node can move when the reference note, result set, model, or indexing state changes. The groups are not a durable taxonomy.
### Troubleshooting weak or noisy Connections
1. Open a meaningful note.
2. Make sure that source import and embeddings are current.
3. Leave auto-refresh running.
4. After major edits, use **Refresh connections**.
5. Examine the exclusions.
6. Tune Sources, Blocks, filters, Hide, Pin, or Pro algorithms only after the basic view returns a useful result.
## Find related notes beside the text you are writing
**Edition: Pro.**
Inline connections use the current paragraph, heading, or block as the semantic anchor and open a focused result popover in the editor.
What the numbers identify:
1. The purple marker belongs to the nearby editor passage.
2. **Lines 3-11** names the block range used as the anchor.
3. Each row names a candidate source.
4. The displayed values are relative connection scores.
*The marker and popover are temporary discovery controls, not persisted links.*
### Enable or disable Inline connections
Run `Smart Connections Pro: Toggle: Inline connections` to show or hide inline result markers.
Assign the command to a hotkey when Inline connections are useful only during review passes.
For an on-demand review pass, draft with Inline off, enable it when you are ready to recover related context or add links, review the useful matches, then disable it when you return to focused writing.
### Understand the inline indicator
A marker appears beside a block when at least one result meets the configured inline threshold.
The marker identifies block-scoped related material. It is not a persisted link.
#### Why do some paragraphs have no inline marker?
No candidate passed the inline score threshold, the block is excluded, or inline processing is disabled for that content.
#### Does the marker create a permanent link?
No. Open or drag the useful result into the note when the relationship should become durable.
### Open the Inline connections popover
Hover the marker to open the related-result popover for that block.
The popover shows the strongest visible candidates and a path to the full list.
### Preview and open a result
Hold Cmd/Ctrl while hovering a result to use Obsidian Hover Preview on desktop. Click a result to open its note.
### Open the full result list for a block
Select the popover header that includes the Connections label and line range. The Connections view opens the full result set for that block.
#### How do I see more than the popover results?
Open the popover header to send that block anchor to the full Connections view.
### Tune the inline score threshold
Inline inherits Connections candidate scoping and adds controls for how much signal appears while editing:
| Control | Effect |
| --- | --- |
| **Inline score threshold** | Sets how strong a match must be before a marker appears. |
| **Results limit** | Sets the maximum number of items shown in the inline popover. |
| **Include and exclude filters** | Restrict candidates by file path where available. |
If normal Connections results work but inline markers do not appear, lower the inline threshold. If the popover is too broad or noisy, reduce its result limit or narrow the candidate paths before changing indexing or model settings.
### Code blocks and unsupported content
Inline processing can skip fenced codeblocks so generated markers do not interfere with code, prompts, or structured block content.
The setting applies only to inline markers. Normal Connections results remain available elsewhere.
### Inline vs Footer vs the Connections view
Use the surfaces as follows:
- Use Inline when one exact block is the intended target.
- Use Footer for note-end review.
- Use the Connections view for persistent whole-note exploration.
What the numbers identify:
1. **Show inline connections** enables paragraph-level markers.
2. The inline threshold controls when a marker appears.
3. **Skip code blocks** controls whether code content receives markers.
4. **Footer connections list component** chooses the note-end renderer. The enable toggle is visible above it.
### Troubleshooting Inline connections
- If normal Connections works but no indicators appear, lower the threshold.
- If the editor is noisy, raise the threshold.
- Make sure that the block is eligible.
- Reopen the Inline connections popover after material edits.
## Show related notes at the bottom of an Obsidian note
**Edition: Core and Pro.**
Footer connections mount the Connections result display at the end of the current note and appear when the note ending is visible.
What the numbers identify:
1. The note content ends before the Footer surface begins.
2. **Smart Connections** identifies the Footer result surface.
3. The expanded row keeps its source identity and score.
4. The preview exposes source content without changing the note.
*Use the footer to inspect note-end results. Expand a result when you need more source content. Placement and gestures can vary on mobile.*
### Enable Footer connections
Run `Smart Connections: Toggle: Footer connections`, or enable Footer connections in Connections settings.
Footer uses the same related-result model as the Connections list, mounted at the end of the editor.
### Choose the Footer list component
**Footer connections list component** controls only the result component mounted at the bottom of notes.
- **List only** keeps the note ending compact and puts the result rows first.
- **Version 4.0 (Graph + List)** adds the Connections mini graph above the same result list. Use it when the visual overview is worth the additional vertical space and rendering work.
This setting is independent from the main **Connections List Component** setting and does not change Connections codeblocks.
There is no separate **Show graph** setting. Select **List only** to omit the graph or **Version 4.0 (Graph + List)** to include it.
#### Why is there no graph in the footer?
The Footer has no graph when **Footer connections list component** is set to **List only**. Select **Version 4.0 (Graph + List)** when you want the mini graph.
### When the footer appears
The footer renders only when the end of the note is visible.
This keeps Footer connections at the end of the note instead of occupying space throughout the document.
#### Why is the footer not visible?
Confirm Footer connections are enabled and scroll until the last line of the note is visible.
### Expand or collapse the footer
Use the footer header to collapse or expand the results.
The footer remembers its folded state.
#### Does collapsing the footer disable Connections?
No. It changes only the footer presentation. Re-expand it to see the same note-level results.
### Read and expand Footer connections results
Expand a result to inspect more context before opening the source.
Expansion does not create a link or change the current note.
### Drag to create a link
On desktop, drag a result from the footer into the editor to insert an Obsidian link.
### Open a result
Open a result when the inline preview is not sufficient.
### Use Footer connections on mobile or without a sidebar
Footer connections provide a note-end surface when a sidebar is impractical. On mobile, use the controls available in your installation.
Use **List only** to keep the Footer compact. Choose **Version 4.0 (Graph + List)** only when the additional overview is worth the extra space.
If Smart Environment loading is deferred:
1. Select **Load Smart Environment**.
2. Wait for **Smart Environment ready** or **Ready**.
3. Then diagnose the Footer.
#### When should I use the sidebar instead?
Use the sidebar for continuous exploration, **Pause auto-refresh**, copying links, and sending a result set to Smart Context.
### Footer vs Inline vs the Connections view
Use the surfaces as follows:
- Use Footer for note-end review.
- Use [Inline connections](https://smartconnections.app/docs/connections/#find-related-notes-beside-the-text-you-are-writing) when one block is the intended target.
- Use the [Smart Connections view](https://smartconnections.app/docs/connections/#find-related-notes-in-obsidian-with-smart-connections) for persistent whole-note exploration.
### Troubleshooting Footer connections
1. Make sure that the setting is enabled.
2. Scroll to the end of the note.
3. Expand the Footer.
4. Make sure that the current note produces normal Connections results.
If the Footer is too large, set **Footer connections list component** to **List only**. If you want the mini graph, select **Version 4.0 (Graph + List)**.
## Tune Smart Connections scoring and ranking
**Edition: Pro.**
These Connections Pro controls shape results in three layers: candidate scope, primary scoring, and final ranking.
### Set the candidate type before scoring
Select **Sources** for note-level candidates or **Blocks** for heading/block-level candidates.
Candidate selection happens before scoring. Fix an overly broad pool before tuning the score.
What the numbers identify:
1. **Connection results type** chooses Sources or Blocks.
2. **Results limit** bounds the visible candidate count.
3. Sidebar location controls where the persistent view opens.
4. **Connections List Component** chooses the current renderer.
*Candidate type and limit shape the pool before scoring or ranking.*
### Configure result presentation
**Edition: Pro.**
- **Connections list item** selects the renderer used for each result row.
- **Show full path** includes the folder path in displayed results.
- **Render markdown** renders result content as Markdown. Turn it off to show plain text instead.
These controls change presentation. They do not change candidate eligibility, scoring, or ranking.
### Configure candidate filters
Filters decide which candidates are eligible.
Use include and exclude rules to set structural boundaries. Examples include project paths, archives, templates, and tags. Apply these rules before you add a more complex scoring algorithm.
What the numbers identify:
1. The filter notes distinguish display filtering from Smart Environment ingestion.
2. **Exclude inlinks (backlinks)** removes notes that already link to the target from displayed results.
3. **Exclude outlinks** removes notes already linked from the target from displayed results.
4. **Include filter** restricts displayed results to paths containing **Newsletter Launch**.
Additional Pro controls in the same settings group:
- **Exclude filter** removes results whose path contains any listed fragment. It runs before **Include filter**.
- **Frontmatter include filter** accepts newline-delimited matchers such as `status` or `status:open`. Frontmatter keys and values are matched case-insensitively.
- **Frontmatter exclude filter** removes matching results. Exclude entries take precedence over frontmatter include entries.
- **Hide frontmatter blocks in results** omits frontmatter-derived Blocks from displayed results so the list shows Sources instead of frontmatter blocks.
*Connections filters change the visible candidate set after Smart Environment builds the dataset. They do not delete notes or change which sources Smart Environment indexes. To stop indexing content, change the applicable Smart Environment include or exclude settings.*
### Choose a score algorithm
Score algorithms compute the primary relevance value.
What the numbers identify:
1. The scoring selector controls the primary relevance calculation.
2. **Cosine Similarity** is selected.
3. The menu lists the current feedback and key/frontmatter alternatives.
4. Ranking remains a separate step below scoring.
*Choose one algorithm. Refresh the same target. Compare the results. Keep the change only when the results improve consistently.*
| Algorithm | Use |
| --- | --- |
| Cosine Similarity | Stable baseline without feedback. |
| Similarity Adjusted by Feedback | Penalize candidates similar to hidden results. |
| Similarity Weighted by Feedback | Boost pinned-like signals and dampen hidden-like signals. |
| Similarity Weighted by Key + Frontmatter | Apply path, key, heading, or metadata multipliers. |
### Choose a ranking algorithm
Ranking algorithms reorder already-scored candidates.
What the numbers identify:
1. Scoring remains **Cosine Similarity**.
2. The separate Ranking algorithm control changes final order.
3. **None** preserves score order.
4. **Re-ranking model** and **Recency rank** are the other visible choices.
*Select a ranking method. Refresh the same target. Compare the results. Keep the change only when the results improve.*
| Algorithm | Use |
| --- | --- |
| None | Preserve score order. |
| Re-ranking model | Apply a configured Smart Rank model to top candidates. |
| Recency rank | Make modification time dominate final order. |
### Start from a known preset
| Goal | Candidate type | Score | Ranking |
| --- | --- | --- | --- |
| Stable note-level baseline | Sources | Cosine Similarity | None |
| Reduce recurring noise after deliberate hides | Sources | Similarity Adjusted by Feedback | None |
| Use stable pinned and hidden preferences | Sources | Similarity Weighted by Feedback | None; add a reranking model only if useful candidates remain out of order. |
| Emphasize paths, headings, or metadata | Blocks | Similarity Weighted by Key + Frontmatter | None; use Recency rank only when freshness should dominate. |
Keep the target, filters, and result limit stable while testing one preset.
#### What is the difference between scoring and ranking?
Scoring decides the primary relevance value. Ranking changes the final order after scoring.
### Use feedback-aware scoring
Feedback-aware algorithms use pinned and hidden signals.
Use them only when those actions represent stable preference. A one-off hide can otherwise create an unintended recurring bias.
### Weight paths, keys, headings, and frontmatter
Weight configuration multiplies the base score when key fragments or metadata match.
```json
{
"key_weights": {
"Projects/": 1.2,
"Readwise/": 0.8
},
"meta_weights": {
"status:evergreen": 1.15,
"type=spec": 1.1
}
}
```
- `key_weights` applies substring matches to source keys, paths, or headings.
- `meta_weights` accepts `key` to match the presence of a frontmatter key, and `key:value` or `key=value` to match a key/value pair.
- Multiple matching weights multiply together.
Use weights for structural intent. Use Recency rank for freshness intent.
### Configure a reranking model
Re-ranking requires a configured Smart Rank model in Smart Environment Pro. Configure the default ranking model in Smart Environment Pro settings before relying on this mode.
When **Re-ranking model** is selected, **Show original connection score** can display the original score alongside the re-ranked score. This changes visible score detail, not model availability or result order.
If no compatible ranking model is available, use None or Recency rank.
### Rank by recency
Recency rank reorders the scored candidates by modification time.
Use it when freshness should dominate. Do not use it to solve an eligibility or semantic-scope problem.
### Troubleshoot broad, weak, or noisy results
| Symptom | First adjustment |
| --- | --- |
| Results are too broad | Lower result limits, add include filters, exclude archives, templates, or exports, or switch to Blocks when long notes hide useful sections. |
| Blocks are too noisy or expensive | Return to Sources or tune block preparation in Smart Environment. |
| The same low-value notes keep returning | Hide them, test Similarity Adjusted by Feedback, or add path or frontmatter excludes. |
| Valuable tagged or foldered notes are under-ranked | Add key or frontmatter weights and verify metadata spelling while keeping other settings stable. |
| Useful candidates are present but ordered poorly | Add one ranking algorithm and compare against the baseline. |
| No useful results appear anywhere | Stop tuning algorithms and verify note eligibility, source preparation, and vault coverage in [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/). |
Change one layer at a time: scope, then filters and limits, then scoring, then ranking.
#### Why did changing three settings at once make results harder to evaluate?
Scope, score, and ranking changed simultaneously. Return to a baseline and alter one layer at a time.
#### Should I use Dedupe when similar notes keep appearing?
Use Dedupe when repeated material needs a keep, merge, archive, or ignore decision. Use algorithm tuning when the issue is result ordering.
### Tune in the correct order
1. Choose Sources or Blocks.
2. Narrow the candidate pool.
3. Establish Cosine Similarity as the baseline.
4. Change scoring only when relevance is consistently wrong.
5. Add ranking only when the correct candidates are in the wrong order.
### Compare algorithm changes
Keep the target, candidate pool, and filters stable. Test one change. Compare useful results across several representative notes. Do not compare score ranges alone.
Do not judge an algorithm from its settings alone. Compare result sets while keeping the target, scope, filters, and candidate pool fixed.
## Rank Obsidian Bases rows by semantic relevance
**Edition: Pro.**
Connections in Bases compares each represented note with a reference note and exposes the result as formula values that can be sorted or filtered.
### Requirements
Use this integration after Smart Connections returns useful results. The active file must be a `.base` file and Connections Pro must expose the score and link-list functions. Both functions appear in Bases formula autocomplete when the integration is active.
### Add a Connections score column
1. Open a `.base` file.
2. Run `Smart Connections Pro: Add: Connections score bases column`.
The command opens a reference-note selector and inserts a formula column that scores each row relative to the selected note.
What the numbers identify:
1. A `.base` file with five reviewed source rows is active.
2. The full **Add: Connections score bases column** command is entered.
3. The matching command is ready to run.
*Run the command while the intended Base is active. Make sure that the new score column appears. Then configure its reference.*
> [!TIP] If the command does not appear
> 1. Make sure that Connections Pro is active.
> 2. Make sure that the Base contains a supported note-file column.
> 3. Reopen the Base.
> 4. Search the Command Palette again.
#### Why does the score-column command not appear?
The active file must be a `.base` file and the Pro Bases integration must be available.
### Choose a fixed reference note
Choose one reference:
- Choose a fixed note for a stable comparison.
- Choose **Current/active file (dynamic)** for a Base that follows the active note.
Dynamic references work best when the Base is kept in the sidebar.
What the numbers identify:
1. The same five-row Base remains visible behind the selector.
2. The requested reference path is visible in the selector.
3. **Landing Page Messaging.md** is the fixed note ready to choose.
#### Why do scores differ from another Base?
Candidate filters, reference note, embedding model, and scoring algorithm can differ. Compare scores only inside the same configured view.
### Use the current active note as a dynamic reference
The dynamic option uses the current active file as the comparison target.
Keep the Base visible in a sidebar when it should behave like a filtered, sortable Connections dashboard.
What the numbers identify:
1. The `.base` file remains the destination for the generated column.
2. The selector is filtered to the active-file option.
3. **Current/active file (dynamic)** is ready to choose.
*Choose a dynamic reference. Change the active note. Make sure that the scores update as expected.*
### Use `score_connection`
`score_connection` returns a numeric connection score using the configured Connections scoring algorithm.
```txt
score_connection(file, file2)
file.score_connection(file2)
```
### Use `list_connections`
`list_connections` returns a list of related links prefixed with their connection scores for a row.
```txt
list_connections(file)
file.list_connections()
```
Use it beside a score column when the Base should show both ranking and an actionable link trail.
#### Should I use `score_connection` or `list_connections`?
Use the score for sorting and filtering. Use the link list when you also want immediate navigation to related notes.
### Bases formula examples
Fixed-reference examples:
```txt
file.score_connection("+Projects/Project Alpha.md")
score_connection(file, "+Projects/Project Alpha.md")
```
Related-link examples:
```txt
file.list_connections()
list_connections(file)
```
### Compare multiple anchors and prepare a shortlist
Add one score column per anchor note when a collection must be evaluated across multiple lenses, such as Goal A, Goal B, Draft, or Constraints. Sort by one column, scan the others, then filter to rows that perform well across the lenses that matter. Review outliers before deciding.
For a grounded AI handoff, review a small shortlist, send only the selected notes to Smart Context, and remove noise before copying or sending the package.
### Sort and filter by relevance
Filter the Base first to define the candidate set, then sort the score column descending.
A score is most useful as a relative ranking inside one Base and one reference point.
What the numbers identify:
1. The Base still contains the same five reviewed rows.
2. The **Relevance** formula column contains Connections values.
3. The fixed reference note scores `1` against itself.
4. The remaining values are ordered from highest to lowest.
*Compare values only inside this Base and reference configuration.*
### Use a dynamic Base in the sidebar
A Base using the active-file reference can stay docked while you move through notes.
The table then reranks only the rows allowed by the Base filters.
### Interpret Connections scores in Bases
Treat the score as a relative signal inside the same Base and reference point. Model, candidate scope, filters, and algorithm choice can change the range. When the configured scoring algorithm uses feedback, pinned or hidden signals can affect values the next time the Base recalculates.
### Troubleshooting Connections in Bases
1. Make sure that a `.base` file is active.
2. Check source preparation.
3. Check the reference note.
4. Dock dynamic Bases in the sidebar.
5. Filter broad collections.
6. Check the configured score algorithm.
## Related documentation
- [Search by a typed question](https://smartconnections.app/docs/lookup/#search-obsidian-notes-by-meaning-with-smart-lookup)
- [Review repeated content](https://smartconnections.app/docs/dedupe/#find-duplicate-and-near-duplicate-notes-in-obsidian)
- [Copy a Base as AI context](https://smartconnections.app/docs/context/#use-bases-as-a-source)
---
## Faq
canonical: https://smartconnections.app/smart-graph/faq/
html_url: https://smartconnections.app/smart-graph/faq/
markdown_url: https://smartconnections.app/smart-graph/faq.md
llms_url: https://smartconnections.app/smart-graph/faq/llms.txt
last_modified: 2026-09-08T17:19:04.418Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Graph FAQs What is Smart Graph for Obsidian?#Smart Graph maps indexed notes into semantic regions so you can review related sources and act on a deliberate selection. The Whole vault view shows the current cluster landscape before you narrow, select, or act. Cluster position and score are current exploration signals, not permanent metadata.What should I do first?#Use this sequence: Open a…
suggested_links:
- title: Getting Started
url: https://smartconnections.app/smart-graph/getting-started/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
# Smart Graph FAQs
## What is Smart Graph for Obsidian?
[Smart Graph](https://smartconnections.app/smart-graph/) maps indexed notes into semantic regions so you can review related sources and act on a deliberate selection.
The Whole vault view shows the current cluster landscape before you narrow, select, or act. Cluster position and score are current exploration signals, not permanent metadata.
## What should I do first?
Use this sequence:
1. Open a meaningful indexed note.
2. Run `Smart Graph Pro: Create graph from current note`.
3. Make sure that the title reads **Smart Graph - Neighborhood of _note name_**.
4. Select one related note.
5. Choose **Open**.
6. Inspect the opened note.
## Which command opens Whole vault?
Run `Smart Graph Pro: Open: Smart Graph view` or use the **Open Smart Graph** ribbon action. Run `Smart Graph Pro: Refresh Smart Graph` when the current view needs to be rebuilt from the index.
## What does Smart Graph require?
Graph needs eligible Smart Sources and current embeddings. A current-note Neighborhood also requires the active note to be indexed. Use **Inspect active note** or **Environment stats** when Graph reports that sources are missing.
## When should I use Smart Graph instead of Smart Connections?
Use [Smart Connections](https://smartconnections.app/smart-connections/) for a ranked list around the active note. Use Graph when spatial regions, a wider scope, or a multi-source action makes the relationship easier to inspect.
## When should I use Smart Graph instead of Smart Lookup?
Use [Smart Lookup](https://smartconnections.app/smart-lookup/) when you start with a phrase or question. Use Graph when you want to explore a semantic area and choose sources visually.
## When should I use Obsidian Search instead?
Use Obsidian Search when you need an exact word, title, heading, tag, operator, syntax pattern, or regular expression. Use Smart Graph when you want to explore a topic visually or review relationships among a selected set of notes.
## How is Smart Graph different from Obsidian's native graph?
Obsidian's native graph centers authored links. Smart Graph starts from semantic similarity. It can also overlay authored links when **Show links** is active.
## What do the lines mean?
Subdued semantic bridges connect current cluster representatives. They are generated exploration signals. **Show links** adds authored source-link edges with arrowheads for direction. Use **Hide links** to remove that overlay.
## Are semantic regions and bridges permanent topics or authored links?
No. Regions, representatives, bridges, positions, and affinities are temporary clues based on the current notes, index, model, scope, and layout. They can change after edits or refresh. Authored links are separate and appear only when the link overlay is enabled.
## Are the nodes real notes I can act on?
Source nodes represent indexed notes or other supported sources. Select a node. Confirm its title in the Notes panel. Choose **Open**. Inspect the source. Cluster regions are computed groupings for exploration. They are not files or permanent folders.
## Does Smart Graph help if my notes are not heavily linked?
Yes. Smart Graph maps semantic similarity, so it can reveal neighborhoods even when authored links are sparse.
## What can I do after I find a useful cluster?
After you find a useful cluster, you can:
- **Open** sources.
- Use **Copy selection links** or **Copy scope links**.
- Create a note or synthesis note.
- Create a Canvas.
- Create a Smart Context.
- Change scope.
- Save a graph image.
Check the opened, pasted, or created destination before you call the action complete.
With one selected source the note action reads **Create note**. With two or more it reads **Create synthesis note**.
## What do Focus cluster, Focus nodes, and Neighborhood mean?
**Focus cluster** renders one semantic region. **Neighborhood** renders a local graph around a source or region. **Focus nodes** renders an explicit selected-source scope and requires at least three selected sources when invoked manually.
Read the view title and Notes panel after changing scope.
*The **Focus nodes** action limits the graph to the selected sources. Regions, positions, and affinities can change after edits, reindexing, or refresh.*
## Does Select matches change the Graph scope?
No. Search changes visual emphasis, **Select matches** changes the selection, and Scope actions change the rendered source set. These states are independent.
## Why is my current note missing?
The note may be excluded, too short under the current source policy, not imported yet, or missing current embeddings. Use this sequence:
1. Choose **Inspect active note**.
2. Resolve the reported source state.
3. Re-import the note.
4. Refresh Graph.
## Why did a note or cluster move?
Graph is recomputed from the current index, embedding model, scope, and note content. Preserve durable conclusions in normal note links or metadata, not cluster numbers or positions.
## What should I do when Graph is empty?
- Prepare indexed Smart Sources when Graph says it needs them.
- Choose **Reset scope** when the current scope returned no cluster data.
- Clear Search when nonmatches are only dimmed.
- Run `Smart Graph Pro: Refresh Smart Graph` after source or embedding changes.
- Reopen the Graph tab after a canvas-rendering failure.
## Does Create canvas preserve Graph groups?
No. The created Canvas can contain selected file cards plus each selected cluster's representative file card even when that representative was not selected; it uses available Graph positions or colors but does not add Canvas group boxes.
## How do I get access to Smart Graph?
Smart Graph is a Pro plugin. Use this sequence:
1. Open the live Smart Plugins Store.
2. Check Smart Graph availability for your account.
3. Follow the visible **Install** or **Enable** action.
4. If **Enable** appears after installation, choose it.
5. Open the Graph workflow with one of the commands above.
## Is Smart Graph experimental?
The current Store lists Smart Graph in the main Pro catalog, not under **Experimental**. The live Store is the authority if that grouping changes. Rows under **Experimental** carry the notice that they are available to Pro subscribers and may change quickly.
---
## Getting Started
canonical: https://smartconnections.app/smart-graph/getting-started/
html_url: https://smartconnections.app/smart-graph/getting-started/
markdown_url: https://smartconnections.app/smart-graph/getting-started.md
llms_url: https://smartconnections.app/smart-graph/getting-started/llms.txt
last_modified: 2026-09-08T17:18:52.063Z
usage_notes: |-
Use this page to answer questions about Getting Started.
excerpt: |-
Getting started with Smart Graph The shortest useful Graph workflow starts from a note you already understand and ends by opening one related note. When to use Smart Graph Use Smart Graph to explore how several notes relate. You can narrow the visible scope or turn a reviewed selection into another artifact. Start with a note you understand so the first relationship is easy to judge. Before you…
suggested_links:
- title: Faq
url: https://smartconnections.app/smart-graph/faq/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
# Getting started with Smart Graph
The shortest useful Graph workflow starts from a note you already understand and ends by opening one related note.
## When to use Smart Graph
Use Smart Graph to explore how several notes relate. You can narrow the visible scope or turn a reviewed selection into another artifact. Start with a note you understand so the first relationship is easy to judge.
## Before you start
Confirm that:
- Smart Graph Pro is installed and active
- the current note is indexed by Smart Environment
- Smart Environment is ready
If the note is missing from Environment stats or Source Inspector, fix that before judging Graph.
## First win: open one related note
### 1. Open a meaningful note
Use a note with a clear topic and enough text to be indexed.
### 2. Create a graph from that note
Run `Smart Graph Pro: Create graph from current note`.
The view title should read **Smart Graph - Neighborhood of _note name_**. The seed note appears in the Notes panel under **Selection**.
### 3. Choose one related note
Select a nearby node whose title looks useful. If reading labels in the canvas is difficult, use this sequence:
1. Choose **Search nodes**.
2. Enter a known title fragment.
3. Choose **Select matches**.
The Notes panel remains headed **Selection** while individual sources are selected. It shows their names and remove controls. Remove the seed if you want **Open** to launch only the new note.
### 4. Open and verify
Choose **Open**. Make sure that the expected note opens in Obsidian. Make sure that its content is relevant to the starting note.
You now have a complete result: current note -> source-seeded Neighborhood -> reviewed related note -> opened source.
Continue with [entry commands and source-seeded behavior](https://smartconnections.app/docs/graph/#open-smart-graph).
> [!NOTE] Preserve conclusions outside the graph
> Graph is an exploratory view. Cluster position, representative notes, and scores can change after edits, indexing, model changes, or scope changes. Record durable conclusions in normal notes and links.
## Search a larger graph
Run `Smart Graph Pro: Open: Smart Graph view` or use the **Open Smart Graph** ribbon action to reveal the existing Smart Graph view when one is open; if none exists, it opens a dedicated root tab in **Whole vault**.
Open Search. Enter one concrete topic, such as `Newsletter Launch`.
Search highlights matches and dims nonmatches. It does not change the selection until you choose **Select matches**.
## Review a selection
The Notes panel keeps selected source identities readable. Treat scores as review signals, not guarantees.
Check:
1. the selected count
2. the visible source identities
3. the highlighted graph nodes
4. the remove controls for noisy sources
Remove anything you would not knowingly open, copy, or send into another workflow.
> [!NOTE] Search, selection, and scope are different
> Search changes emphasis. Selection chooses sources for an action. Scope changes which sources Graph renders.
For selection gestures and panel behavior, see [Select sources and clusters](https://smartconnections.app/docs/graph/#select-sources-and-clusters) and [Use the Notes panel](https://smartconnections.app/docs/graph/#use-the-notes-panel).
## Focus the reviewed nodes
Choose **Scope** -> **Focus nodes** to make the selected sources the visible Graph scope.
The view title changes to **Smart Graph - Focus nodes**. When the selection is cleared, the Notes panel is headed **Focus nodes scope** and lists the visible source set. Manual **Focus nodes** requires at least three selected sources.
The **Focus nodes** title and source list confirm the five-source scope. The open search remains a separate filter inside that scope.
## Turn a reviewed selection into something useful
The dock offers **Open**, **Copy**, **Create**, and **Scope** when their grouped menus are needed; a group with only one visible action renders that action directly.
Choose an action. Inspect its destination before you treat the result as complete.
The action bar offers:
- **Open** to inspect selected sources
- **Copy selection links** to copy Obsidian links
- **Create** to choose **Create note** or **Create synthesis note**, **Create canvas**, **Create context from selection**, or a graph image
- **Scope** to focus the graph or inspect a neighborhood
After you choose an action, check its destination. Inspect the opened notes, pasted links, created artifact, or sources transferred to Context Builder.
For confirmation rules and output-specific caveats, see [Use Smart Graph actions](https://smartconnections.app/docs/graph/#use-smart-graph-actions).
## Find every primary Graph surface
| Surface | Open it | Ready when | Learn more |
| --- | --- | --- | --- |
| Current-note Neighborhood | `Smart Graph Pro: Create graph from current note` | The view title names the seed. The seed appears under **Selection**. | [Entry commands](https://smartconnections.app/docs/graph/#open-smart-graph) |
| Whole vault | `Smart Graph Pro: Open: Smart Graph view` | The view title reads **Smart Graph - Whole vault**. | [Scope modes](https://smartconnections.app/docs/graph/#choose-a-graph-scope) |
| Search | Choose **Search nodes**. | Matches are highlighted and **Select matches** is available. | [Search behavior](https://smartconnections.app/docs/graph/#search-the-graph) |
| Notes panel | Select a node, region, or search result set. | The panel is headed **Selection**, **Selected cluster**, or **Selected clusters**, with source names and review controls below it. | [Selection review](https://smartconnections.app/docs/graph/#use-the-notes-panel) |
| Action dock | Keep at least one source selected. | **Open**, **Copy selection links**, **Create**, and **Scope** controls are visible. | [Actions and outputs](https://smartconnections.app/docs/graph/#use-smart-graph-actions) |
| Authored links | Choose **Show links** from Graph actions. | Directed source-link arrows are visible separately from semantic bridges. | [Semantic and authored edges](https://smartconnections.app/docs/graph/#understand-visible-bridges) |
## If the first graph is confusing
| Symptom | First adjustment |
| --- | --- |
| Too many nodes | Search for one recognizable topic or use a smaller scope. |
| Search highlights the wrong area | Use a more specific term. Review the matches before you select them. |
| Selection contains noise | Remove individual sources before using an action. |
| `Smart Graph needs indexed Smart Sources before it can render.` | Open Source Inspector or Environment stats. Prepare eligible sources. |
| `Current note is not indexed by Smart Environment.` | Check the active note in Source Inspector. Re-import it or remove its exclusion. |
| `This scope did not return any cluster data.` | Choose **Reset scope**. |
| Search looks empty | Clear the query. A zero-match search does not delete nodes. |
| Canvas did not render | Run `Smart Graph Pro: Refresh Smart Graph`. Reopen Graph. |
| A group moved after refresh | Treat groups as recomputed current-run signals, not saved categories. |
| You need one related note | Use Connections instead. |
| You have a question to type | Use Lookup instead. |
| You need an exact word, title, heading, tag, operator, syntax pattern, or regex | Use Obsidian Search instead. |
## Continue from here
- [Use all Graph actions](https://smartconnections.app/docs/graph/#use-smart-graph-actions)
- [Interpret cluster signals safely](https://smartconnections.app/docs/graph/#interpret-clusters-representatives-and-affinities)
- [Build a reviewed Smart Context package](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian)
---
## Graph
canonical: https://smartconnections.app/docs/graph/
html_url: https://smartconnections.app/docs/graph/
markdown_url: https://smartconnections.app/docs/graph.md
llms_url: https://smartconnections.app/docs/graph/llms.txt
last_modified: 2026-09-08T16:04:39.706Z
usage_notes: |-
Use this page to answer questions about Graph.
excerpt: |-
Smart Graph Smart Graph maps indexed notes into a semantic landscape. Start from one current note when you want a related note. Open Whole vault when you want a wider pattern. New to Smart Graph? Start with Getting started with Smart Graph. Create a Neighborhood from the current note. Then open one related note. Explore semantic clusters in Obsidian with Smart Graph Smart Graph presents indexed…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
# Smart Graph
Smart Graph maps indexed notes into a semantic landscape. Start from one current note when you want a related note. Open Whole vault when you want a wider pattern.
> [!TIP] New to Smart Graph?
> Start with [Getting started with Smart Graph](https://smartconnections.app/smart-graph/getting-started/). Create a Neighborhood from the current note. Then open one related note.
## Explore semantic clusters in Obsidian with Smart Graph
Smart Graph presents indexed source notes as a semantic landscape. Clusters, representatives, and affinities are current hypotheses for inspection, not permanent authored categories.
The Whole vault view establishes the current Graph surface and cluster landscape. Narrow it before you act. Clustering and scores are exploration signals, not final judgments.
### Before opening Smart Graph
Smart Graph requires eligible Smart Sources and current embeddings. To start from the active note, that note must be indexed. Use **Inspect active note** or **Environment stats** when Graph reports that the note or source set is unavailable.
### Open Smart Graph
Choose the entry that matches the task:
| Command | Result |
| --- | --- |
| `Smart Graph Pro: Create graph from current note` | Opens a **Neighborhood** seeded by the indexed active note and selects that seed. |
| `Smart Graph Pro: Create graph from selected note` | Opens a source-seeded graph for the selected indexed note. |
| `Smart Graph Pro: Open: Smart Graph view` | Opens **Whole vault**. The **Open Smart Graph** ribbon action uses this route. |
| `Smart Graph Pro: Refresh Smart Graph` | Rebuilds the current Graph view from the current index. |
Foreground nodes represent indexed source notes. Blocks resolve to their owning notes rather than appearing as separate foreground nodes.
What to notice:
1. The active tab identifies the Smart Graph workspace.
2. **Smart Graph - Whole vault** states the current scope.
3. Search is available before narrowing or selecting the graph.
Use the view title to confirm whether you opened **Whole vault** or **Neighborhood of _source_**.
### Choose a graph scope
Smart Graph supports:
- **Whole vault**
- **Neighborhood**
- **Focus cluster**
- **Focus nodes**
The current scope is shown in the view title after **Smart Graph -**.
*Compare a focused source scope with the same vault's broader graph.*
### Add notes and folders by dropping them into Smart Graph
Drop indexed notes, supported folders, Connections or Lookup results, or a graph source onto Smart Graph.
- Whole vault becomes a **Focus nodes** scope containing the dropped indexed sources.
- An existing focused scope keeps its current members, adds newly dropped sources, and becomes a **Focus nodes** scope.
- A dropped folder expands recursively to indexed source notes.
- A dropped block resolves to its owning indexed source note.
The resolved dropped sources become the current selection when the scope changes. A duplicate-only drop leaves the scope unchanged. Unindexed items and ambiguous paths are rejected.
### Understand source nodes
Foreground nodes represent indexed source notes. Visual emphasis can reflect stronger members or the current representative without changing the note's identity.
### Understand cluster regions
Semantic regions sit behind source nodes and can be selected as a group. They are current clustering results, not permanent topic objects.
#### Are semantic regions permanent topics?
No. They can change after note edits, reindexing, reclustering, model changes, or scope changes.
### Understand visible bridges
Smart Graph can show two different edge types:
- Semantic bridges connect current cluster representatives. They are generated from the current index and scope. Treat them as exploration signals.
- Authored source links appear after **Show links**. They come from note links, use arrowheads for direction, and can show arrows in both directions.
Use **Hide links** to remove authored-link edges. Do not describe a semantic bridge as an authored link, or an authored link as proof of semantic similarity.
### Select sources and clusters
- Click a source to add it to the selection.
- Alt/Option-click removes only that source.
- Drag across the graph to select by area.
- Click a semantic region to add its members.
- Click the background to clear the selection.
- Hold Space while dragging to pan without changing selection.
- While holding Space, use the scroll wheel to zoom in or out.
- Mod-click a source to open it while preserving the previous selection.
### Use the Notes panel
The Notes panel changes its visible heading with its contents:
- **Selection** for an ordinary source selection
- **Selected cluster** or **Selected clusters** when the selection contains complete semantic regions
- **Focus nodes scope**, **Focus cluster scope**, or **Neighborhood scope** when no source is selected and the panel is listing an explicit scope
Hover a row to highlight its node. Select a row to move the camera to that source without changing the graph selection. Use the row remove control to subtract a selected source.
### Search the graph
The Newsletter Launch query highlights matching sources while preserving the surrounding landscape.
What to notice:
1. The query defines the current visual search filter.
2. **Select matches** adds the visible matches to the selection.
3. Matching nodes are emphasized but are not selected until committed.
4. Nonmatches remain visible in a dimmed surrounding landscape.
Search matches source labels and full source paths.
- Matches are emphasized without being selected.
- Nonmatches are dimmed.
- Selected nonmatches remain visible.
- A zero-match query leaves the graph usable instead of hiding everything.
- **Select matches** adds all current matches to the selection.
- Clear or press Escape to close search while preserving the committed selection.
#### Does selecting a search match change the graph scope?
No. Search, selection, and scope are separate. Use **Select matches** or a scope action explicitly.
### Use graph context menus
Right-click behavior depends on the target:
- source: include that source in the selection and show relevant actions
- semantic region: include its members and show relevant actions
- background or visible bridge: preserve the current selection and show graph-wide actions
### Camera and viewport behavior
A graph rebuild refits the rendered scope so its visible elements return to view.
The camera button in the lower-right corner saves the current graph image to the vault root with a name shaped like `smart-graph-YYYY-MM-DD-hh-mm-ss.png`. After saving, Graph attempts to copy the same image to the clipboard for pasting. Verify the saved file and clipboard result separately.
#### Why did the camera refit after refresh?
A graph rebuild refits the current rendered scope.
### Empty and error states
| Message | Recovery |
| --- | --- |
| `Smart Graph needs indexed Smart Sources before it can render.` | Prepare eligible sources in Smart Environment. Refresh Graph. |
| `Current note is not indexed by Smart Environment.` | Inspect the active note. Resolve the reported exclusion or source-eligibility problem. Re-import the note. |
| `This scope did not return any cluster data. Reset scope to show the whole vault.` | Choose **Reset scope**. |
| `Unable to build Smart Graph cluster data.` | Confirm source and embedding readiness. Run `Smart Graph Pro: Refresh Smart Graph`. |
| `Smart Graph canvas was not ready to render.` | Refresh or reopen the Graph tab. |
### Use Smart Graph actions
The actions overlay and right-click menus provide the same selected-source, scope, creation, and save actions.
#### Actions overlay
The actions overlay exposes operations for the current source selection and the current graph scope. Controls appear only when their required selection or scope is available.
Its primary controls are **Open**, **Copy**, **Create**, and **Scope** when their grouped menus are needed; a group with only one visible action renders that action directly. Open **Create** to choose an artifact or **Scope** to change the rendered source set.
What to notice:
1. The selected-source panel shows the reviewed source that remains active while the menu is open.
2. **Open** acts on the reviewed source.
3. **Copy selection links** copies the reviewed source identity.
4. **Create** exposes note, Canvas, graph-image, and Context choices. **Scope** remains a separate boundary.
After you use an action, check its destination. Make sure that the applicable output matches the reviewed selection or scope. The output can be opened notes, pasted links, a created note or Canvas, a saved image, or sources transferred to Context Builder.
#### Open
**Open** launches the selected source notes. For a large selection, read and confirm any warning before opening many notes at once.
##### Why did Open ask for confirmation?
More than ten sources were selected. The second click confirms the large open action.
#### Copy selected links
Use **Copy selection links**. Paste the result into a clean note. Make sure that every intended source appears as an Obsidian wikilink.
#### Copy scope links
Use **Copy scope links**. Paste the result into a clean note. Make sure that the links match the current explicit Graph scope.
#### Add the selection to Smart Context
Choose **Create context from selection**. Confirm every transferred source in Smart Context Builder. Then copy or save the package.
*The same three source identities remain inspectable from selection through the native Create action to Context Builder. This sequence shows a reviewed handoff, not a copied or sent package.*
[[+Outcome/screenshot/sequence/assets/graph-selection-to-context-review-editorial-16x9-dark-graph-v1.2.1_context-v3.4.1.png|Open full-size sequence]]
#### Add the scope to Smart Context
If **Create context from scope** is unavailable, check Smart Context in **Browse Smart Plugins**. Install it if it is not installed. Enable it if it is not enabled. Then choose **Create context from scope**. Make sure that Builder contains every intended source from the explicit Graph scope.
#### Create a note
With one selected source, choose **Create note**. With two or more selected sources, the label becomes **Create synthesis note**. Inspect the new note. Depending on the selection, it can include:
- current scope
- selected-source count
- editable synthesis prompts
- source links grouped by current semantic group
- possible bridge candidates when the graph contains them
Group and bridge suggestions are review aids, not conclusions.
*The generated note remains editable after Graph creates it from the selected sources.*
##### Does Create note infer a final topic name for each group?
No. It creates an editable synthesis scaffold from the current graph.
#### Create a Canvas
Choose **Create canvas**. Inspect the new Canvas. Make sure that every selected source appears. Cards can use Graph positions and colors when available. Adjust the layout before you treat it as a durable artifact.
The generated Canvas does not add Canvas group boxes.
*The created Canvas includes every selected file card and may also include each selected cluster's representative file card even when that representative was not selected. It uses available Graph placement; Graph groups are not converted into Canvas group boxes.*
##### Does Create canvas add Canvas group boxes?
No. It arranges file cards spatially without adding group boxes.
#### Save graph image
Choose **Save graph image** or use the lower-right camera button. Open the timestamped PNG in the vault root. Make sure that it contains the intended viewport, scope label, selected-source markers, and a readable source key. Paste the clipboard image separately and compare it with the saved output.
*This is the saved output itself: selected-node markers and the source key remain visible without live Graph controls.*
#### Focus nodes
**Focus nodes** changes the graph to the selected-source scope. The manual action is available when at least three sources are selected.
After the scope changes, the view title reads **Smart Graph - Focus nodes**. Clear the selection to review the full focused set under **Focus nodes scope** in the Notes panel. Treat the topic interpretation and node positions as current exploration results.
What to notice:
1. **Focus nodes scope** lists the five sources now visible in the graph.
2. The `Newsletter` query remains a search filter inside the focused scope.
3. A matching source stays emphasized without changing the scope.
4. Visible relationships are recomputed Graph signals, not new authored links.
#### Focus cluster
**Focus cluster** limits the graph to the selected semantic group's members.
#### Neighborhood
**Neighborhood** creates a local graph around a source or selected semantic group.
#### Reset scope
**Reset scope** returns a scoped graph to **Whole vault**. It is hidden while Whole vault is already active.
##### Why is Reset scope missing?
The graph is already in Whole-vault scope, or the active view does not have a narrower scope to reset.
#### Right-click menus
- Right-click a source or semantic region for selection actions and graph-wide actions.
- Right-click the graph background or a visible bridge to keep the current selection and open graph-wide actions.
- With no selection, background and bridge menus omit selection-only actions.
#### Replay and link controls
Graph-wide actions can include **Show links** or **Hide links**, **Replay graph reveal**, and **Replay semantic cascade**. Replay actions animate the existing view. They do not change source content or create a durable artifact.
### Interpret clusters, representatives, and affinities
Clustering describes the current index and scope. Identifiers and affinities can change after edits, reindexing, or another run.
#### Whole-vault clustering
Whole-vault scope groups the currently indexed notes by semantic similarity. The groups are recomputed from the current index rather than stored as permanent folders or tags.
#### Scoped clustering
Focused scopes cluster only the source notes included in that scope. Block-based entry points resolve to their owning notes before graph construction.
#### Neighborhood clustering
Neighborhood scope uses the selected source or cluster as a seed so the local graph stays centered on that starting point.
#### Cluster representative
Each group uses one current note as its representative. The representative is the note nearest the group's semantic center. It is not an authored title or permanent category name.
##### Is the representative note the official group name?
No. It is only the current note nearest the group center.
#### Group members
A note's strength inside a group is relative to the current clustering run. Use it for visual emphasis and review, not as a durable score to store in notes.
#### Relationships between groups
Smart Graph ranks current relationships between groups. These signals can support layout, visible bridges, and bridge candidates in generated notes, but they remain hypotheses for inspection.
#### Interpret clustering scores
Compare scores only within the same graph generation and scope. Reindexing, changing models, editing notes, or changing scope can change membership and score ranges.
#### Durability limits
Do not treat current group numbers, positions, representatives, or bridge scores as permanent identifiers. Use normal note links or durable note metadata when a relationship must persist beyond the current graph.
##### Why did a note move to another group?
The graph was recomputed from a changed index, scope, model, or note content.
##### Can I save a group number as a permanent label?
No. Current group identities are not durable.
### Large-vault and performance guidance
Use a bounded scope when Whole vault is dense or slow. Start with `Create graph from current note`. You can also use Search, Focus nodes, Focus cluster, or Neighborhood. A ready view or a smaller scope is not a speed benchmark.
### Troubleshooting Smart Graph
1. Read the Graph view title to identify the current scope.
2. Clear Search if nonmatches are merely dimmed.
3. Read the Notes panel heading: **Selection** describes selected sources, while a heading ending in **scope** describes the visible scope.
4. Choose **Reset scope** when a focused scope is empty.
5. Check Source Inspector or Environment stats for missing sources or embeddings.
6. Run `Smart Graph Pro: Refresh Smart Graph`.
7. If rendering still fails, reopen the tab.
### Requirements and availability
This page documents Smart Graph Pro on desktop. The current Store lists Smart Graph in the main Pro catalog rather than under **Experimental**. Use the live Store as the authority if that grouping changes. Context creation actions require Smart Context. Graph image, note, Canvas, and Context actions should be considered complete only after their destination opens and matches the reviewed selection or scope.
## Related documentation
- [Explore a current-note result list](https://smartconnections.app/docs/connections/#find-related-notes-in-obsidian-with-smart-connections)
- [Review a Smart Context source package](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian)
---
## Named
canonical: https://smartconnections.app/smart-context/builder/named/
html_url: https://smartconnections.app/smart-context/builder/named/
markdown_url: https://smartconnections.app/smart-context/builder/named.md
llms_url: https://smartconnections.app/smart-context/builder/named/llms.txt
last_modified: 2026-09-08T13:27:16.143Z
usage_notes: |-
Use this page to answer questions about Named.
excerpt: |-
Named contexts Deprecated page This page is retained temporarily and may contain outdated details. Use the current named-context documentation for current behavior. Start with Getting started with Smart Context for the shortest verified workflow. Named contexts are saved Smart Context bundles you can reopen, copy, and refine without rebuilding the same selection every time. Build it once. Name it…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
# Named contexts
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current named-context documentation](https://smartconnections.app/docs/context/#smart-context-named-contexts) for current behavior. Start with [Getting started with Smart Context](https://smartconnections.app/smart-context/getting-started/) for the shortest verified workflow.
Named contexts are saved Smart Context bundles you can reopen, copy, and refine without rebuilding the same selection every time.
Build it once. Name it once. Reuse it anywhere the same job comes back.
## Save a named context
Build a working set in Smart Context Builder, then type a name in the header.
As soon as the context has a name, it becomes a saved Smart Context that you can reopen later.
Name the context bundles however you like.
Slash names also group cleanly in the dashboard list, like "nested tags" in Obsidian.
## Open the named contexts dashboard
Use the named contexts dashboard to browse saved contexts and reopen them in the Builder.
From the dashboard, you can:
- open a saved context back in the Builder
- copy it directly to the clipboard
- delete it when it is no longer useful
That makes the dashboard the home base for reusable context packs.
## Add items by dropping them onto a saved context
Drop notes, supported media, folders, Connections results, or Lookup results onto an existing row. Folders expand recursively to supported context items.
Named-context-to-named-context drops remain unavailable until cycle handling is explicit. Add a named context through Builder when one saved context should reference another.
## Drag a named context into Smart Chat or a note
Drag the row header into Smart Chat to add the saved bundle as a named-context reference. Drag the same header into a supported Markdown editor to insert a complete canonical `ctx` codeblock.
## Common workflow
1. Build a working set in the Builder.
2. Give it a name.
3. Reopen it from the dashboard the next time the same problem returns.
4. Copy it directly, or edit it in the Builder before copying.
Named contexts work especially well for:
- recurring project briefs
- meeting prep packs
- writing voice packs
- bug triage bundles
- client handoff sets
> [!NOTE] Watch a named context become reusable
> [](https://youtu.be/_i3577ti8jg?t=675)
>
> Callum shows a saved context as a reusable bundle that can be reopened, reused, and kept lean as the project changes.
## Named context vs codeblock
Use a named context when:
- the bundle should be reused across many notes
- the selection is a repeatable working set
- you want one saved item in the dashboard
Use a codeblock when:
- the context should stay attached to one specific note
- the note itself is the brief, prompt, or control surface
- the manifest should stay visible in the note body
## Related pages
- [Build and save reusable context packs](https://smartconnections.app/smart-context/builder/)
- [Copy notes and folders as AI-ready context](https://smartconnections.app/smart-context/clipboard/)
- [Attach a Smart Context codeblock to a note](https://smartconnections.app/smart-context/codeblock/)
- [Use file navigator actions to copy notes and folders as context](https://smartconnections.app/smart-context/file-nav-actions/)
- [Copy the current note with link-depth control](https://smartconnections.app/smart-context/clipboard/current/)
- [Control Smart Context templates and export format](https://smartconnections.app/smart-context/settings/)
---
## Context
canonical: https://smartconnections.app/docs/context/
html_url: https://smartconnections.app/docs/context/
markdown_url: https://smartconnections.app/docs/context.md
llms_url: https://smartconnections.app/docs/context/llms.txt
last_modified: 2026-09-07T21:29:26.872Z
usage_notes: |-
Use this page to answer questions about Context.
excerpt: |-
Smart Context Smart Context turns notes and other supported sources into a package you can inspect before copying, exporting, or sending it. The first current-note copy workflow is available in Core. This page shows the complete Pro interface. Requirements and availability summarizes the edition boundaries. New to Smart Context? Start with Getting started with Smart Context. The first workflow…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
- title: Graph
url: https://smartconnections.app/docs/graph/
# Smart Context
Smart Context turns notes and other supported sources into a package you can inspect before copying, exporting, or sending it.
The first current-note copy workflow is available in Core. This page shows the complete Pro interface. [Requirements and availability](https://smartconnections.app/docs/context/#requirements-and-availability) summarizes the edition boundaries.
> [!TIP] New to Smart Context?
> Start with [Getting started with Smart Context](https://smartconnections.app/smart-context/getting-started/). The first workflow copies one active note at Depth 0 and verifies the pasted result.
What the numbers show:
1. **Copy context** is the action to use after reviewing the package.
2. The expanded source tree shows what the package currently contains.
3. The source-mode tabs choose how to find more evidence.
4. Search results are candidates until you add them.
## In this guide
- [Copy reviewed context](#smart-context-clipboard)
- [Start from the current note](#smart-context-current-note)
- [Build reusable context packages](#smart-context-builder)
- [Inspect and recover Context rules](#smart-context-builder-rules)
- [Save and reuse named contexts](#smart-context-named-contexts)
- [Attach a context manifest to a note](#smart-context-codeblock)
- [Add external files](#smart-context-external-files)
- [Copy images and PDF pages](#smart-context-media)
- [Use Canvas and Bases as sources](#smart-context-canvas)
- [Customize copied text](#smart-context-output-templates)
- [Use Smart Context on mobile](#smart-context-mobile)
- [Recover from missing or oversized sources](#smart-context-recovery)
## Copy reviewed context
Choose the smallest output that fits the task. A completed copy notice shows that the action ran. The pasted or opened result shows whether the content is correct.
| Output | Edition | What it puts on the clipboard | Verify |
| --- | --- | --- | --- |
| **Copy text** | Core and Pro | A text package with a context tree and source items. | Paths, source bodies, and exclusions match the reviewed package. |
| **Copy media** | Pro | One composite clipboard image built from selected images and rendered PDF pages. | Every intended image or page appears and remains readable. |
| **Copy ZIP** | Pro | A ZIP file reference. | Open the archive and inspect its file list before sharing it. |
| **Copy link tree** | Core current-note command and Pro package menus | A compact text hierarchy of source links. | The hierarchy contains the intended sources. |
| **Copy with Template** | Core and Pro, with Smart Templates | A template-shaped text result. | The selected template and sources both appear. This action is added by Smart Templates when it is installed and the package contains items. |
If a media or ZIP safety limit appears, cancel and reduce the selection. Use **Zip anyway** only after reviewing the warning and destination.
### Choose depth-based or direct copy
Only current-note modal copy opens the estimated **Copy context** depth chooser. Use it when the active note is the traversal root and you need to compare depth and backlink rows.
The following actions are direct and do not open that chooser:
| Starting surface | Package boundary |
| --- | --- |
| A fixed-depth current-note command | Uses the depth and backlink mode named by the command. |
| Selected notes or folders in Files | Uses the eligible selected sources. |
| One folder from its context menu or the folder picker | Uses the folder's eligible contents. |
| Context Builder | Uses the reviewed Builder package as it currently resolves. |
| A Smart Context codeblock | Quick copy uses the resolved codeblock package. Its menu can apply a selected depth without opening the current-note chooser. |
| A named-context dashboard row | Uses the saved package as it currently resolves. |
Use a direct action only when the package boundary is already correct. Use Builder when sources need review, removal, or naming before copy. **Copy text** and **Copy media** are separate clipboard actions; when a task needs both, copy and verify each output separately.
### File navigator actions
Right-click a supported selection in Obsidian's Files view, then choose the action that matches the amount of review you need:
| Action | Use it when | What happens |
| --- | --- | --- |
| **Copy selected notes as context** | At least two selected Markdown notes already form the right package. | Copies them immediately. No link-depth chooser opens. |
| **Copy selected folders as context** | At least two selected folders already form the right package. | Resolves their eligible notes into one direct package. |
| **Copy selection as link tree** | A mixed selected set should be represented as a compact hierarchy of Obsidian links. | Copies the selected hierarchy without opening the link-depth chooser. |
| **Copy folder contents to clipboard** | One folder already defines the right package. | Resolves that folder's eligible contents and copies them immediately. No link-depth chooser opens. |
| **Copy folder as link tree** | One folder should be represented as a compact hierarchy of Obsidian links. | Copies the folder hierarchy without opening the link-depth chooser. |
| **Open selection in Context Builder** | Selected notes or folders need review, removal, naming, or another output. | Opens the selection as a reviewable Builder package. |
| **Open folder in Context Builder** | One folder needs review before copy. | Opens the folder-derived package in Builder. |
| **Select folder to copy contents** | A searchable folder picker is faster than navigating the Files view. | Resolves the chosen folder and copies its eligible contents immediately. No link-depth chooser opens. |
Use direct copy when the selection is already correct. Use Builder when you must remove items, name the package, choose another output, or examine Pro Rules. Direct file and folder actions do not use a selectable link depth. Use the current-note workflow when the task requires link traversal.
Use a link-tree output only when the destination needs an outline of source links rather than the full note text, and when that option is available.
## Start from the current note
Core and Pro expose different first-copy controls:
- **Core:** From the Command Palette, run `Smart Context: Copy current text to clipboard (choose link depth)`. In the **Copy context** modal, select the **Depth 0** row marked **Current note**. Selecting the row copies immediately.
- **Pro:** Choose the **Smart Context: Copy current** ribbon action. The menu begins with **Choose link depth...** and direct depth choices. Open **Depth 0 - current note**. Then choose **Copy text**.
Pro also provides `Smart Context Pro: Copy current context clipboard (choose link depth)` in the Command Palette. It opens the estimated **Copy context** modal. Selecting a row copies immediately.
What the numbers show:
1. In the Pro menu, **Choose link depth...** opens the estimated modal.
2. **Depth 0 - current note** opens the direct output submenu for the active note and material it embeds.
3. **Depth 1 - outlinks only** opens the direct output submenu for outgoing linked notes.
4. **Depth 1 - include backlinks** opens the direct output submenu for incoming and outgoing links.
What the numbers show:
1. In the **Copy context** modal, selecting **Depth 0** with **Current note** copies text immediately.
2. **Depth 1** with **Outlinks only** follows outgoing links.
3. **Depth 1** with **Include backlinks** follows incoming and outgoing links.
4. In Pro, **Shift + Select** copies media for the selected depth.
Depth 0 does not traverse ordinary links. Embedded notes can still appear because they are part of the current note's rendered content. Higher depths follow links breadth-first.
### Choose the smallest useful depth
| Depth | Adds | Use it when |
| --- | --- | --- |
| **0** | The current source, embedded material, and depth-0 codeblock additions. | The active note already contains the assignment and evidence. |
| **1 - outlinks only** | Notes linked from the starting source. | Intentional outgoing links hold the missing evidence. |
| **1 - include backlinks** | Outgoing and incoming linked notes. | Notes that point to the starting source also matter. |
| **2 or 3** | Additional linked layers. | A bounded multi-hop relationship is required and the larger package remains reviewable. |
Fixed-depth commands skip the chooser and work well as hotkeys. Core provides direct copy for depth 0, depth 1, and depth 1 with backlinks. Pro adds depth 2 and depth 3, with and without backlinks.
The first Smart Context codeblock in the active note can add items to the depth-0 package. When those additions change the result, the chooser can show a **No codeblock** row so you can copy the note without them. Embedded notes and a Base embedded in the starting source are also handled at depth 0.
Character, item, and token counts are estimates for choosing a manageable package. The destination can count tokens differently.
For either edition, finish the workflow this way:
1. Open the note that owns the task.
2. Copy at the smallest useful depth using the controls for your edition.
3. Read the copy notice for file and character counts.
4. Paste into a clean note or composer.
5. Make sure that the `(current)` source, every `- ` path, and all embedded material are correct.
6. Widen the depth only when the missing evidence is linked from or back to the note.
In Pro, hold Shift while choosing a depth in the chooser when you intend to copy media instead of text.
## Build reusable AI context packages in Obsidian
From the Command Palette, run the Builder command for your installed edition:
- Core: `Smart Context: Open new context in builder`
- Pro: `Smart Context Pro: Open new context in builder`
In either edition, you can also choose the **Smart Context: Open Builder** ribbon action. Builder separates the package being reviewed from the source mode used to find evidence.
### Read the active package
The Builder header shows **Build context**, **Add sources**, and a review area with:
- the active or saved context name
- source count
- estimated token count
- a hierarchical source tree
- copy and package-management controls
In Pro, the header also shows the Rules count. Token counts are estimates. A destination can tokenize the same text differently.
### Choose among the nine source modes
| Source mode | Edition | Use it for |
| --- | --- | --- |
| Notes | Core and Pro | Markdown notes, Bases, and Canvas files. |
| Media | Pro | Images and PDF attachments that need visual context. |
| Folders | Pro | A dynamic folder-derived source group. Direct folder copy remains available in Core. |
| Sections | Core and Pro | A specific heading or block instead of a whole note. |
| Named contexts | Core and Pro | A saved, reusable source package. |
| Linked notes | Pro | Notes connected to an intentional starting source. |
| Similar notes | Pro | Semantically related candidates that still require review. |
| Tags | Pro | A dynamic group derived from a tag. |
| External files | Pro, desktop | Files or folders outside the vault. |
The search footer shows the available keyboard action. **Enter** adds the highlighted source. `Cmd + Enter` or the right arrow opens available blocks or sections. **Shift + Enter** follows links up to the available depth. Use Left Arrow to return to the parent suggestion list.
Choose **Sections** when only one heading or block is relevant. A smaller section usually produces a more inspectable package than a whole long note.
### Check where each source came from
Keep the tree expanded long enough to verify where each source came from. Direct, section-derived, folder-derived, and named-context-derived items can have different update and exclusion behavior.
For mixed-origin packages, expand each branch and read its origin badge. A named context or folder can expand into more items than its dashboard count because its saved rules are applied when Builder resolves the package.
Leaf rows can show source type, size, share of the package, origin, and missing-source state. Remove or restore a missing source before relying on the result. In Core, edit a source named context when one of its derived items cannot be removed independently.
The current source-row menu lists **Open source**, **View connections**, and **Remove from context** in that order. Use the menu to inspect the source, continue discovery in Connections, or change package membership. Make sure that the intended destination or package change appears after the action.
### Remove, clear, or manage the package
Removing a direct source removes that selection. In Pro, an inherited item can come from a folder or another dynamic group. When you remove that item, Pro can create an exclusion rule. The rule keeps the item out while the group remains included.
After a rename, duplication, clear, deletion, or source removal, reopen the package. Make sure that the intended change appears. **Clear** removes the current package contents. **Delete context** removes a saved context definition.
### Inspect and recover Context rules
**Edition: Pro.**
Open the rule-count control before copying a package assembled from folders, tags, or other dynamic sources.
What the numbers show:
1. The visible `1 rule` chip preserves continuity with the reviewed package.
2. The global `1 exclude` count and override show the current exclusion state.
3. The Template global heading-exclusion row is visibly On for this package.
4. Source-mode tabs remain available while Rules are inspected.
Rules make dynamic packages easy to review:
- **Include** records a dynamic group that contributes sources.
- **Exclude** keeps a matching source or heading out.
- Switching off a global heading exclusion creates an exception for this context. It does not delete the global rule.
- Removing an included group changes the package. It is not the same as restoring one excluded child.
Excluded source rows expose **Restore** for individual recovery. The Rules overview exposes **Restore all excludes** for bulk recovery. Use the narrowest applicable control. These controls change Context rules, not the source notes. Reopen the package and make sure that the intended sources return before copying.
#### Override a global heading rule for one context
1. The context name and source boundary identify the package being edited.
2. `Template` is a global heading exclusion and is On for this package.
1. The package and sources stay the same.
2. Turning `Template` Off creates an exception for this context.
1. Switch the override.
2. Copy the package.
3. Make sure that the excluded heading appears.
4. Restore the rule.
5. Copy the package again.
6. Make sure that the heading is omitted.
7. Make sure that the source remains included.
The override does not change the source note.
With the package exception active, the compiled source includes **Template**, its two reviewed lines, and the following **Review boundary**.
[[context-per-context-template-override-sequence-03-exception-output-includes-template-sanitized-documentation-1280x280-desktop-2026-08-06.png|Open full-size image]]
After the global exclusion is restored, the same source wrapper and **Review boundary** remain while the **Template** section is absent.
*These excerpts verify the pictured heading boundary, not completeness of the whole context package.*
[[context-per-context-template-override-sequence-04-global-output-excludes-template-sanitized-documentation-1280x285-desktop-2026-08-06.png|Open full-size image]]
## Save and reuse named contexts
A named context stores a reviewed source definition. Reopen it to resolve the current context package.
### Save a named context
Enter a clear name in Builder after the source list is correct. In Pro, check Rules too. Use a name that describes the reusable source set. For example, use `Newsletter Launch Evidence` instead of the one-off prompt.
### Reopen and inspect before reuse
Open the named-context dashboard or select **Named contexts** in Builder. Before copying, check the source count, availability, and estimate. In Pro, also check Rules. A saved name does not guarantee that every source is present or current.
What the numbers show:
1. `Newsletter Launch Evidence` is the selected named context whose menu is open.
2. Copy text, media, ZIP, and link-tree choices provide different outputs from the same package. **Copy with Template** is supplied by Smart Templates when available.
3. Builder, copy, clear, and delete choices manage the selected named context.
Use the dashboard to:
- reopen a named context in Builder
- copy its resolved package
- duplicate a named context
- rename a named context
- clear a named context
- delete a named context
**Make a copy**, **Clear this context**, and **Delete context** affect the saved definition. Before you select one of these actions, make sure that the selected name is correct.
After a lifecycle action:
1. Reopen the package.
2. Check its name.
3. Check its source count.
4. Check source availability.
5. In Pro, check Rules.
### Add to and reuse a named context by drag and drop
Drop notes, folders, Connections results, Lookup results, and supported media where available onto an existing named-context row. Folders expand recursively into supported context items. After the drop, verify the updated source count and resolved tree.
Named-context-to-named-context drops are not supported because they can create direct or transitive cycles. When one saved context should reference another, add the referenced named context through **Named contexts** in Builder.
Drag the row header into Smart Chat to add a named-context reference. Drag the same header into a supported Markdown editor to insert a complete canonical `ctx` codeblock.
## Attach an AI context manifest to an Obsidian note
A Smart Context codeblock keeps a source manifest beside the note that uses it. Run the command for your installed edition:
- Core: `Smart Context: Insert codeblock`
- Pro: `Smart Context Pro: Insert codeblock`
Core codeblocks accept local note paths and `ctx::` named-context lines:
~~~~markdown
```ctx
projects/Newsletter Launch/00 - Newsletter Launch Plan.md
projects/Newsletter Launch/04 - Channel Plan.md
ctx:: Newsletter Launch Evidence
```
~~~~
Pro also accepts supported relative external include and exclude lines:
~~~~markdown
```ctx
../example-project/src
!../example-project/dist
!*.test.js
```
~~~~
In Reading view, use the rendered action bar only after the source and rendered tree match. The `ctx`, `context`, and `smart-context` fence aliases render the same Smart Context surface. Rewrites preserve the alias used in the note.
*Open the manifest in Builder when it needs review. Copy text, media, ZIP, or a link tree only after the rendered sources match the fence.*
The menu can also create a named context, open the named-context dashboard, or copy at a selected depth. Pro external include lines and `!` exclusion lines remain visible in the note so the package boundary can be reviewed later.
### Choose the copy scope
Use the codeblock copy action when you want only the package resolved from that codeblock.
Use the regular current-note copy workflow when you want one package containing the active note, sources selected by the current-note depth, and the first Smart Context codeblock's resolved sources. Direct declarations from that codeblock join the current-note package at Depth 0 and do not start additional link traversal by themselves.
### Check the rendered codeblock
Compare the fenced source with the rendered tree. Use an action only after the two views match. After copying or opening the block in Builder, make sure that named contexts resolve to the intended sources. In Pro, make sure that external paths also resolve correctly.
## Add external files and code repositories to Smart Context
**Edition: Pro. Desktop only.**
**External files** adds files and folders outside the vault. Store portable relative paths when possible. Never publish an absolute home path.
*Set **Max search depth** and **Max search items** to bound external discovery. Before you scan a repository, exclude dependencies, generated output, caches, and secrets with ignore patterns.*
The supported syntax includes:
- `../example-project/src` to include a relative source tree
- `!../example-project/dist` to exclude a relative subtree
- `!*.test.js` to exclude matching files
After you add an external path, expand the resolved tree. Make sure that included files are present. Make sure that excluded folders and patterns are absent. External files are desktop-oriented.
Use **Max search depth**, **Max search items**, ignore patterns, and scan limits to keep repositories reviewable. The external-folder item cap is applied to raw scan results per folder expansion: scan-time regex filters run before the cap, while structured exact and folder exclusions run after it, so excluded candidates can prevent later eligible candidates from being considered. Exclude dependencies, builds, caches, generated output, secrets, and unrelated packages before copying. Prefer one broad folder exclusion over many child exclusions when the boundary is stable.
## Copy images and PDF pages as AI context
**Edition: Pro.**
Use **Copy media** for screenshots, diagrams, whiteboards, PDF tables, charts, figures, slide snapshots, and other visual evidence. It places media on the clipboard separately from **Copy text**.
Prepare the sources before copying:
- Embed a required visual in the starting note when it must be available at Depth 0.
- When the note only links to an attachment, use the smallest depth that reaches it.
- The media count reports how many items the selected package and depth resolved.
Use the route that matches the package you are reviewing:
- **Current note:** Use the current-note controls described above. In the Pro menu, choose the smallest useful depth and then **Copy media**. In the estimated depth chooser, **Shift + Select** copies media for the selected depth.
- **Builder or named context:** Review the resolved package, then choose **Copy media** from its copy menu. In Builder, use **Media** to add the intended images or PDF files before copying.
At Depth 0, supported media embedded in the starting note is eligible. Linked attachments are eligible only when they fall within the selected link depth. Use the media count to choose the smallest depth that still includes the intended visuals. Text estimates describe the text package; the media count describes the candidate composite image.
**Copy media** renders supported images and each selected PDF page into one composite clipboard image. Image aspect ratios are preserved, and PDF pages remain attributable through their file and page labels. The clipboard result contains media only. When the task also needs text, run **Copy text** separately and verify both pasted outputs.
After copying:
1. Read the copy notice. Use its media count as a quick check that the intended attachments were found.
2. Paste into the exact destination you plan to use.
3. Confirm that a destination which accepts pasted images shows an image attachment or thumbnail for the composite.
4. Confirm that every intended image and PDF page is present and readable.
5. Check filenames and page labels where the destination displays them.
6. Compare the pasted composite with the reviewed package before sending.
7. If the destination rejects the paste, retry in an empty composer before changing the sources.
8. If the destination is text-only, use **Copy text** instead.
Clipboard capabilities vary by platform and destination, so an app can paste only the format it supports. Smart Context does not fetch images that exist only as external web URLs. If the safety limit appears, reduce the number of items or their resolution.
## Use Canvas as a source
Use the current-note copy workflow while a Canvas file is active. The Notes source mode also includes Canvas files for Builder workflows. Add a Canvas only when its referenced material belongs in the package, then inspect the resolved source tree before copying.
Use a Canvas when you already organize the project visually, want to review relationships before copying, or need a quick snapshot of linked notes. Supported file cards provide the starting sources, and the selected depth determines whether Smart Context also follows links from them.
At Depth 0, the active Canvas contributes sources represented directly by its supported file cards. Higher depths follow linked evidence from those resolved sources. The copied result represents source content, not the Canvas layout: card positions, groups, and edge meaning are not encoded as semantic structure. Use **Copy media** or another visual export when the Canvas arrangement itself is part of the evidence.
*The chooser applies link depth to the Canvas-resolved file cards. Text copy does not preserve card positions, groups, or edge meaning.*
## Use Bases as a source
The Notes source mode includes Bases. Add the Base. Then examine which notes or rows appear in the reviewed tree.
Use depth 0 for the rendered Base itself. Use depth 1 only when notes linked from the rendered output are required. Narrow the Base first: the rows currently visible in the source view are not a guaranteed copy boundary. After copying, inspect which rows, filters, properties, links, and rendered text reached the destination. Relative `this.file` or `this.note` behavior must be judged from the compiled result, not the Base view alone.
Smart Context can compile a `.base` as context. Semantic Connections scoring inside Bases is a separate [Connections Pro](https://smartconnections.app/docs/connections/#rank-obsidian-bases-rows-by-semantic-relevance) capability.
### Add a Base in Builder
1. Open Builder.
2. Choose **Notes**.
3. Search for and add the `.base` file.
4. Expand the context tree to see the resolved rows or notes.
5. Remove unrelated material.
6. Use Depth 0 for the rendered Base. Use Depth 1 only for a known linked note.
If more records appear than expected, return to Depth 0 or add the required notes directly instead of widening the package again.
## Customize copied text
Smart Context settings control how the complete package and each source item are formatted. Current presets include structured XML, Markdown headings, JSON, and custom templates.
*Choose an output preset. Paste a small package. Examine the exact structure. Then use the preset in a larger workflow.*
> [!IMPORTANT] JSON output
> **JSON Stringify** is the Pro serializer-backed output path. Handwritten JSON templates can become invalid when copied content contains quotes, backslashes, newlines, control characters, or trailing delimiters. Test pasted JSON with a parser before you use it in downstream automation.
The context template controls the overall wrapper, including headers, footers, and the optional file tree. The item template controls how each note title or path appears and, where supported, can format whole-note and block items differently.
Long wrappers and a file tree increase copied size. Remove nonessential wrapper content when the package exceeds the destination limit.
The context template can use `{{FILE_TREE}}` for a hierarchical source map. Item templates support:
| Variable | Value |
| --- | --- |
| `{{KEY}}` | Full item key or path |
| `{{ITEM_NAME}}` | File or block name without folder path or extension |
| `{{TIME_AGO}}` | Relative modified time or `Missing` |
| `{{LINK_DEPTH}}` | Link depth, default `0` |
| `{{EXT}}` | File extension when available |
| `{{IS_CURRENT}}` | A marker when the item belongs to the active note |
`{{FILE_TREE}}` marks the first active-note match as current. Use `{{IS_CURRENT}}` when every matching item needs a marker.
### Apply heading filters
**Edition: Pro.**
Heading filters change the copied package, not the source notes. Matching is case-insensitive. When a source has at least one configured Include heading, Smart Context copies only sections whose headings match. Otherwise, it removes sections that match Exclude headings and preserves the remaining content.
Block-only items and external-file items are not sliced by source-note heading filters. Examine a small pasted package before you apply heading filters to a larger workflow.
## Use Smart Context on mobile
In-vault note, section, named-context, codeblock, and text-copy workflows can remain useful on mobile. External file and folder browsing requires desktop filesystem access. A named context created on desktop can still open on mobile, but external items in it cannot resolve there.
Clipboard behavior depends on the mobile operating system, Obsidian version, destination app, and payload size. There is no universal maximum: start with a small depth and verify the actual paste in the destination.
## Recover from missing or oversized sources
| What you see | What to do |
| --- | --- |
| A missing-source row | Remove the row or restore the file. Reopen the context. Check the tree. |
| A folder marked truncated | Narrow the folder, use a Section, or split the package. |
| An incomplete-output confirmation | Cancel the copy. Reduce the selection. Copy again unless the partial package is intentional. |
| A media safety-limit warning | Reduce the number of images or PDF pages, or lower their resolution. |
| An empty copy result | Make sure that the package contains at least one included item. Make sure that Rules did not exclude all items. |
| Clipboard copy fails | Keep Builder open. Retry once. If the copy fails again, copy a smaller package or use ZIP. |
| Current-note package is too broad | Reduce link depth. |
| Folder package is too broad | Use a hand-picked selection or open the folder in Builder. |
| Template overhead is too large | Remove nonessential wrappers and the file tree. |
| One long note dominates the package | Select a Section or block-level item in Builder. |
Character, item, token, and size estimates are approximate; destination limits apply to the actual copied or archived output.
## Privacy and trust
Smart Context assembles the package in Obsidian. Content can leave Obsidian when you copy, paste, export, or send it to another application or provider.
Before sending:
- inspect every source and exclusion
- remove private or irrelevant material
- verify relative paths do not reveal a host identity
- inspect the pasted text or media in the destination
- treat model output as untrusted until reviewed against the sources
## Requirements and availability
This page documents the current desktop Pro interface. The first useful copy does not require Pro.
| Edition | Current boundary |
| --- | --- |
| Core | Current-note copy with the depth chooser.
Direct selected-note and folder copy.
Builder note and block selection.
Named contexts.
Note and named-context codeblocks.
Template-driven exports.
Canvas sources. |
| Pro | Media, images, and PDFs.
External files, folders, and repositories.
Dynamic folder and tag groups in Builder.
Exclusions and heading filters.
Richer Bases rendering and advanced workflows. |
Direct folder copy is Core. A dynamic folder or tag group inside Builder is Pro. Basic `.base` source handling is not a Pro-only claim. Pro adds the richer Bases workflow.
Use the live Smart Plugins Store to see which Context edition is installed and active. **External files** requires desktop file access. **Copy with Template** requires Smart Templates. On mobile, use only the source and output controls visible in that build. Examine the pasted result before you send it.
## Related documentation
- [Getting started with Smart Context](https://smartconnections.app/smart-context/getting-started/)
- [Smart Context FAQs](https://smartconnections.app/smart-context/faq/)
- [Structure a reviewed context as a reusable prompt](https://smartconnections.app/docs/templates/)
- [Find candidate sources with Smart Connections](https://smartconnections.app/docs/connections/)
- [Use reviewed context in Smart Chat](https://smartconnections.app/docs/chat/)
---
## Builder
canonical: https://smartconnections.app/smart-context/builder/
html_url: https://smartconnections.app/smart-context/builder/
markdown_url: https://smartconnections.app/smart-context/builder.md
llms_url: https://smartconnections.app/smart-context/builder/llms.txt
last_modified: 2026-09-07T21:24:29.710Z
usage_notes: |-
Use this page to answer questions about Builder.
excerpt: |-
Smart Context Builder Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Context Builder documentation for current behavior. Start with Getting started with Smart Context for the shortest verified workflow. Smart Context Builder is the fastest way to assemble a reusable context bundle inside Obsidian. Search, add, review, name, and copy from…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-context/bases/
- title: Canvas
url: https://smartconnections.app/smart-context/canvas/
- title: Clipboard
url: https://smartconnections.app/smart-context/clipboard/
- title: Codeblock
url: https://smartconnections.app/smart-context/codeblock/
- title: Faq
url: https://smartconnections.app/smart-context/faq/
# Smart Context Builder
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Context Builder documentation](https://smartconnections.app/docs/context/#smart-context-builder) for current behavior. Start with [Getting started with Smart Context](https://smartconnections.app/smart-context/getting-started/) for the shortest verified workflow.
Smart Context Builder is the fastest way to assemble a reusable context bundle inside Obsidian.
Search, add, review, name, and copy from one native modal.
## Open the Builder
Run:
```obsidian-command
Smart Context: Open Selector for New Context
```
The empty state starts with three entry points:
- Add sources
- Add named contexts
- Add blocks
That lets you start from a whole note, an existing named context, or a precise block.
## Select sources with fuzzy search
Start typing to filter note titles.
Press `Enter` to add the highlighted source.
As soon as you add something, the selected tree stays pinned at the top of the modal so you can keep building without losing track of the current bundle.
### Use blocks when precision matters
Use whole sources when the note is short or mostly relevant.
Use blocks when the note is long and only a section matters.
From a highlighted source:
- `Enter` adds the source
- `Right Arrow` opens block suggestions
- `Ctrl/Cmd + Enter` opens block suggestions
## Review the current selection
The Builder keeps your current context visible above the suggestion list.
This gives you a quick read on:
- the current tree
- live character and token estimates
- which items are already in the bundle
Use the remove control on any row to trim the set while you build.
> [!NOTE] Watch Builder prune a context bundle
> [](https://youtu.be/_i3577ti8jg?t=717)
>
> This clip shows the judgment step: use related notes to create the bundle, then prune it so only useful sources remain before delegation.
### Remove single items or clear everything
- Remove any individual item with the **x** control (highlighted in pink).
- Use **Clear** to remove all selected items at once.
- Watch the total size details (characters and token estimate) to keep tabs on exported context size.
- Use the per-item size to quickly find the biggest items when you need to shrink the bundle.
## Save as a named context
Type a name in the header to save the current selection as a reusable Smart Context.
This is the fastest way to turn a one-off working set into something you can reopen later from the named contexts dashboard.
Slash names also group cleanly in the named contexts dashboard (like nested tags in Obsidian).
## Copy from the Builder
Use the copy action for a direct text export, or open the menu for the full set of Builder actions.
Current actions shown here:
- `Copy text`
- `Copy link tree`
- `Clear this context`
Use `Copy text` when you want the merged context bundle.
Use `Copy link tree` when you want a compact markdown tree of the selection instead.
## Context item types
Core Builder supports:
- sources
- blocks
- named contexts
Other Core copy surfaces can still copy selected notes and folders directly from the file navigator.
Pro extends the Builder with advanced source types and controls, including:
- folders
- tags
- external files and folders
- exclusions for advanced context groups
- Context rules with per-context overrides for global heading exclusions
That keeps the Core flow simple while still letting power users build larger working sets from repo files, external folders, or dynamic groups.
## Advanced context building
### Dynamic groups (Pro)
Add folders and tags so that the named context stays up-to-date with their contents.
### Rules and exclusions (Pro)
Use Context rules to review the dynamic rules affecting the current Smart Context.
Folder and named-context rules add sources. Exclude rules prevent matching folders, notes, patterns, or headings from being copied or exported. For example, add a code repository and exclude its `packages` folder.
Plugin-level excluded headings also appear under **Global settings**:
- `On` means the heading exclusion applies to this Smart Context.
- `Off` allows that heading in this Smart Context without changing plugin settings.
- Disabled rules remain visible so you can turn them back on.
The Builder saves these choices with the Smart Context. When its heading choices match plugin settings again, the instance override is removed and the context resumes inheriting global heading settings.
### External sources (Pro)
Build contexts using files from outside your Obsidian vault.
## Related pages
- [Copy notes and folders as AI-ready context](https://smartconnections.app/smart-context/clipboard/)
- [Copy the current note with link-depth control](https://smartconnections.app/smart-context/clipboard/current/)
- [Use file navigator actions to copy notes and folders as context](https://smartconnections.app/smart-context/file-nav-actions/)
- [Copy context with images and PDFs](https://smartconnections.app/smart-context/clipboard/media/)
- [Copy context from Obsidian Canvas files](https://smartconnections.app/smart-context/canvas/)
- [Attach a Smart Context codeblock to a note](https://smartconnections.app/smart-context/codeblock/)
- [Reopen and reuse saved named contexts](https://smartconnections.app/smart-context/builder/named/)
- [Control Smart Context templates and export format](https://smartconnections.app/smart-context/settings/)
---
## Smart Plugins
canonical: https://smartconnections.app/smart-plugins/
html_url: https://smartconnections.app/smart-plugins/
markdown_url: https://smartconnections.app/smart-plugins.md
llms_url: https://smartconnections.app/smart-plugins/llms.txt
last_modified: 2026-09-07T19:38:08.091Z
usage_notes: |-
Use this page to answer questions about Smart Plugins.
excerpt: |-
Smart Plugins directory Smart Plugins for Obsidian Start with one useful result from your own notes. Add the next Smart Plugin when you need a specific workflow: discovery, context, structure, durable chat, landscape, cleanup, or ChatGPT connection. Install Smart Connections Explore Smart Plugins Pro Source available | Runs in Obsidian | Providers are opt-in Choose the right starting point…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Smart Plugins directory
# Smart Plugins for Obsidian
Start with one useful result from your own notes. Add the next Smart Plugin when you need a specific workflow: discovery, context, structure, durable chat, landscape, cleanup, or ChatGPT connection.
[Install Smart Connections](https://community.obsidian.md/plugins/smart-connections)
[Explore Smart Plugins Pro](/pro-plugins/)
Source available | Runs in Obsidian | Providers are opt-in
## Choose the right starting point
**Current note** -> Connections
**Question** -> Lookup
**Exact phrase** -> Obsidian search
**Landscape** -> Graph
**Reusable set** -> Context
**Repeatable form** -> Templates
**Durable thread** -> Chat
**Approved action** -> Connect Pro
Flagship Core
## Smart Connections
Related notes surface while you write.
- - Current-note related note discovery
- - Preview, drag, copy, or send strong matches to Context
[Find related notes with Smart Connections](/smart-connections/)
Core
## Smart Lookup
Ask naturally. Retrieve by meaning.
- - Use when the question is the anchor
- - Expand results and confirm relevance before you act
[Search notes by meaning with Smart Lookup](/smart-lookup/)
Core
## Smart Context
Build, copy, save, and reuse the context your AI work needs.
- - Copy notes, folders, or blocks in one step
- - Save named packs for repeated work
[Build reusable AI context with Smart Context](/smart-context/)
Core
## Smart Templates
Good outputs need both context and structure.
- - Build prompts from the current note or selection
- - Reuse trusted Markdown templates as output forms
[Reuse Markdown templates as AI prompts](/smart-templates/)
Core / Pro
## Smart Chat
Keep AI conversations where the work lives.
- - Note-based chat threads and bookmarks
- - Pro adds advanced routing when model control matters
[Keep AI chat attached to notes](/smart-chat/)
Pro
## Smart Graph
See the shape of a topic, then act on the notes inside it.
- - Map clusters and neighborhoods by meaning
- - Select notes visually, then open, copy, or add to Context
[Explore Smart Graph semantic vault maps](/smart-graph/)
Pro
## Connect Pro
Turn an action you approve in chat into vault changes you can review.
- - Connect intentionally and run approved actions
- - Review the command result before repeating the workflow
[Turn ChatGPT into vault actions with Connect Pro](/connect-pro/)
Upgrade
## Smart Plugins Pro
Core gives you the shortest path to value. Pro adds advanced workflows when your work needs more control.
- - More context, faster retrieval, model routing, generation, cleanup, and graph visualization
- - One all-access trial across available Pro workflows
[Unlock advanced Smart Plugins Pro workflows](/pro-plugins/)
---
## Smart Templates
canonical: https://smartconnections.app/smart-templates/
html_url: https://smartconnections.app/smart-templates/
markdown_url: https://smartconnections.app/smart-templates.md
llms_url: https://smartconnections.app/smart-templates/llms.txt
last_modified: 2026-09-07T19:38:08.091Z
usage_notes: |-
Use this page to answer questions about Smart Templates.
excerpt: |-
Reusable AI prompts for Obsidian Smart Templates for Obsidian: Reuse your Markdown templates as AI prompts Stop rebuilding the same prompt every time you use AI. Start from the note or selection you care about, apply one or more templates you already trust, and get a repeatable prompt for meeting follow-ups, research briefs, summaries, and outlines. Pro adds generate inside Obsidian. Install from…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Reusable AI prompts for Obsidian
# Smart Templates for Obsidian: Reuse your Markdown templates as AI prompts
Stop rebuilding the same prompt every time you use AI.
Start from the note or selection you care about, apply one or more templates you already trust,
and get a repeatable prompt for meeting follow-ups, research briefs, summaries, and outlines.
Pro adds generate inside Obsidian.
[Install from Community Plugins](https://obsidian.md/plugins?id=smart-templates)
[Explore Templates Pro](/pro-plugins/)
Works with your existing templates | Current note or selection as context | Core: no API setup | Pro: generate inside Obsidian
Core
### Build the prompt and copy it
Best when you use ChatGPT, Claude, Gemini, Smart Chat, or another chat UI and just want a better prompt.
Pro
### Generate, review, and insert
Best when you want to stay in Obsidian, review output, then insert it or create a note.
[Install + first win](#first-win)
[Obsidian Templates vs Smart Templates](#obsidian-templates-vs-smart-templates)
[Proof](#proof)
[Core vs Pro](#core-vs-pro)
[Workflow fit](#workflow-fit)
[FAQ](#faq)
One shared v2 flow: start from context, select templates, then copy a prompt in Core or generate in Pro.
## Install Smart Templates and get your first useful prompt in under a minute
Install the plugin, open a note you care about, and turn that note plus one template into a repeatable AI prompt.
1. Step 1
### Open the shared modal
Install Smart Templates from Community Plugins, then run Smart Templates: Open template context from the command palette or ribbon.
2. Step 2
### Start from the current note or selection
The current note is the context anchor. If you highlight text first, the selection becomes the starting context.
3. Step 3
### Choose a template and run it
Pick one template, add a short instruction, then click Copy prompt in Core or Generate in Pro.
### First-win use cases
**Meeting notes:** turn raw notes into action items, decisions, and follow-ups.
**Research notes:** turn highlights into summaries, open questions, and next steps.
**Drafts and outlines:** keep your preferred structure instead of re-explaining it every time.
## Obsidian Templates vs Smart Templates
Obsidian Templates inserts predefined note content. Smart Templates uses note context plus templates to build a prompt, and Pro can generate inside Obsidian.
### Obsidian Templates
- ✓Insert predefined content into notes.
- ✓Great for note scaffolding and repeatable note creation.
- ✓Does not package the current note as AI context.
### Smart Templates
- ✓Starts from the current note or selection.
- ✓Reuses existing template files, flags, filename rules, or template headings.
- ✓Builds a prompt in Core, or generates and reviews output in Pro.
Already use the Obsidian Templates folder? Point Smart Templates at the same source in
[Smart Templates settings](/smart-templates/settings/)
.
## Why Smart Templates works
It closes the gap between what your notes already know and what your prompt usually forgets.
### Start from context, not a blank box
The current note is the default anchor, and a text selection can replace it before you choose a template.
Learn the
[Template Context modal](/smart-templates/modal/)
.
### Detect templates where they already live
Use configured folders, the Obsidian Templates folder, a smart template: true flag, a filename rule, or matching headings.
Configure detection in
[Settings](/smart-templates/settings/)
.
### Use one shared flow for Core and Pro
Core exposes Copy prompt. Pro adds Generate plus a review modal for copy, insert, or create-note actions.
Compare
[clipboard](/smart-templates/clipboard/)
and
[generate](/smart-templates/generate/)
workflows.
### Built-in defaults remove setup friction
Even before you configure vault-backed sources, Smart Templates includes defaults such as summary, tags, research paper, and diagram.
## Pick the workflow that matches your friction
Smart Templates pays off when the structure repeats even when the raw material changes.
### Meeting follow-up
Turn raw meeting notes into action items, owners, risks, and decisions.
### Research brief
Start from a source note or excerpt, add supporting context, then apply a briefing template.
### Writing structure
Keep your preferred outline, voice constraints, or section order without restating them every time.
### Multi-template output
Combine templates when one structure is not enough, such as a recommendation table plus an executive summary.
## Where Smart Templates fits in the Smart Plugins workflow
Use Smart Templates when structure is the problem. Use the rest of the ecosystem when discovery, context assembly, or ongoing chat is the bottleneck.
[Need better context
### Smart Context
Assemble the right notes, folders, or named packs before you apply a template.](/smart-context/)
[Best downstream pair for Core
### Smart Chat
Paste copied prompts into chats that stay attached to the right notes.](/smart-chat/)
[Need discovery
### Smart Connections
Find related notes before packaging them into a structured prompt.](/smart-connections/)
## Proof you can inspect right now
The best current proof is what you can inspect immediately: the plugin listing, the settings guide, and release notes.
Install
Obsidian Community plugin
Review the plugin entry and install Smart Templates from the official Obsidian directory.
[Inspect
Settings guide
See the live detection rules for folders, naming conventions, and heading-based discovery.](/smart-templates/settings/)
[Shipped changes
Release notes
See the shipping trail for context handling, prompt copy, review, and output actions.](/smart-templates/releases/)
## What stays under your control
- ✓
Core is clipboard-first. Your notes stay in Obsidian unless you choose to paste the copied prompt into another tool.
- ✓
Your templates stay in Markdown inside the vault, where they remain inspectable, editable, and shareable.
- ✓
Pro only sends content when you explicitly choose Generate and configure a model provider.
## Smart Templates Core vs Templates Pro
Start with Core for clipboard-first prompt building. Add Pro when you want to generate, review, and insert results inside Obsidian.
Core plugin
### Smart Templates Core
Free, context-first prompt building from the templates you already keep in your vault.
- ✓
Open one shared modal from the current note, selection, ribbon, or file menu.
- ✓
Add context, select one or more templates, and copy the final prompt.
- ✓
Reuse vault templates plus built-in defaults, with no API setup.
Pro plugin
In-Obsidian generation
### Templates Pro
Everything in Core plus generation, output review, insert, create note, and model override.
- ✓
Generate from the same shared modal instead of switching to an external chat.
- ✓
Review output, then copy, insert into the current note, or create a new note.
- ✓
Set a default model in settings, then override it per request.
## Guides and next steps
Start here
Getting Started with Smart Templates
The fastest path from one note to one repeatable prompt.
[Core loop
Clipboard workflow for ChatGPT, Claude, Gemini, and more
Build a prompt from context plus templates and copy it for any external chat.](/smart-templates/clipboard/)
[Shared UX
Template Context modal
See the context-first runtime that powers both Core and Pro.](/smart-templates/modal/)
[Detection rules
Settings for folders, filenames, and headings
Configure where templates are found and how Pro behaves.](/smart-templates/settings/)
[Entry points
Commands, ribbon, and file-menu entry points
Open the shared modal from the command palette, ribbon, or file menu.](/smart-templates/commands/)
[Pro workflow
Generate inside Obsidian
Generate, review, insert, and create notes without leaving the vault.](/smart-templates/generate/)
## FAQ
Quick answers on compatibility, Core vs Pro, privacy, and fit.
What is Smart Templates for Obsidian?
Smart Templates turns the Markdown templates you already trust into reusable AI workflows. It starts from your current note or selection, adds context, then packages that context with one or more templates so you can copy a prompt or generate inside Obsidian with Pro.
Does Smart Templates replace Obsidian Templates?
No. Obsidian Templates inserts predefined note content. Smart Templates uses those same templates as AI structure layered on top of the current note or selection.
Do I need an API key?
Not for Core. Core builds prompts and copies them to your clipboard so you can use ChatGPT, Claude, Gemini, Smart Chat, or another chat interface. Pro adds in-Obsidian generation.
What is the difference between Smart Templates Core and Templates Pro?
Core covers template discovery, context-first prompt building, multiple template selection, and clipboard export. Pro keeps the same shared modal, then adds Generate, review, insert, create note, and model override.
How are templates detected?
Smart Templates can discover templates from configured folders, the Obsidian Templates folder as a fallback, notes flagged with smart template: true, a configured filename match, or matching headings. Built-in defaults stay available even before setup.
Can I keep multiple templates in one note?
Yes. Matching headings can act as templates, so you can keep multiple reusable template blocks inside one note.
Can I use more than one template at once?
Yes. Smart Templates can merge multiple templates in selection order when one output needs more than one reusable structure.
Can I use Smart Templates with ChatGPT, Claude, Gemini, or Smart Chat?
Yes. Core is clipboard-based, so the same prompt can be pasted into ChatGPT, Claude, Gemini, Smart Chat, or another chat interface. Pro adds a native generate path.
How is Smart Templates different from Smart Context?
Smart Context assembles the right notes as grounded context. Smart Templates applies reusable output structure to that context. They work well together when you need both.
Does Smart Templates keep my notes private and local-first?
Core is clipboard-based: your notes stay in Obsidian unless you paste the copied prompt into another tool. In Pro, content is only sent when you explicitly choose Generate and configure a model provider.
What is the fastest way to get started?
Install Smart Templates from Community Plugins, open a note you already care about, run the shared modal, choose one template, add a short instruction, then click Copy prompt. The [Getting Started guide](/smart-templates/getting-started/) walks through that first win.
Keep your best structures in Markdown, package them with the right context, and stop rebuilding prompts from scratch.
[Install from Community Plugins](https://obsidian.md/plugins?id=smart-templates)
[See Getting Started](/smart-templates/getting-started/)
---
## Smart Lookup
canonical: https://smartconnections.app/smart-lookup/
html_url: https://smartconnections.app/smart-lookup/
markdown_url: https://smartconnections.app/smart-lookup.md
llms_url: https://smartconnections.app/smart-lookup/llms.txt
last_modified: 2026-09-07T19:38:08.087Z
usage_notes: |-
Use this page to answer questions about Smart Lookup.
excerpt: |-
Question-first semantic search for Obsidian Smart Lookup for Obsidian: search notes by meaning, not exact words You know the note exists, but you do not remember the phrase. Ask your vault a plain-language question and surface relevant notes by meaning so wording drift does not break your flow. Problem Keyword guesses, filename hunting, and tab hopping slow you down when recall is fuzzy. Turning…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Question-first semantic search for Obsidian
# Smart Lookup for Obsidian: search notes by meaning, not exact words
You know the note exists, but you do not remember the phrase. Ask your vault a plain-language question and surface relevant notes by meaning so wording drift does not break your flow.
Problem
Keyword guesses, filename hunting, and tab hopping slow you down when recall is fuzzy.
Turning point
Treat search like a question, not a keyword puzzle. Smart Lookup ranks notes by semantic similarity to your query.
Outcome
Find the right note fast enough to keep writing, reuse prior thinking, and act on the result immediately.
Plain-language queries | Meaning-ranked results | Same semantic layer as Smart Connections
[See the search workflow](/smart-lookup/search/)
[Compare with Connections](/smart-connections/)
[First win](#first-win)
[When to use Lookup](#when-to-use)
[How it works](#how-it-works)
[Next step](#ecosystem-fit)
[FAQ](#faq)
Want the annotated walkthrough? Open the
[Smart Lookup search guide](/smart-lookup/search/)
.
## Get your first useful answer in under a minute
Start with one real question you would ask yourself if the right note were already open.
1. Step 1
### Open Smart Lookup
Open the Lookup view from the command palette, the ribbon, or a hotkey you assign.
2. Step 2
### Ask one plain-language question
Describe the idea, topic, or decision you want to recover. Add 1-2 context nouns when your vault is broad.
3. Step 3
### Confirm and act
Expand 1-2 results, open the strongest match, then link it, build a reading trail, or export it into Smart Context.
### What counts as first value?
**Recovered note:** you find something that exact keyword search would have made harder to recover.
**Faster confirmation:** the right heading, block, or note appears quickly enough that you stay in flow.
**Next action:** the result becomes a link, a reading trail, or grounded AI context instead of another dead end.
Once it earns a permanent place in your workflow, pin a hotkey so question-first search stays one move away.
## When Smart Lookup beats keyword search
Use Lookup when the problem is recall, wording drift, or concept discovery rather than exact phrase matching.
### Fuzzy recall
You remember the idea, not the phrase, file name, or heading you used when you wrote it.
### Vocabulary drift
Your current wording changed, but the underlying idea is still in the vault.
### Decision recall
You want to recover tradeoffs, prior decisions, or the best note on a topic before you re-solve it.
### Discovery before drafting
You want a fast reading trail from your prior thinking before you write, delegate, or brief an AI tool.
### Choose the right surface fast
Exact match
#### Obsidian Search
Best when you need exact phrases, tags, operators, file names, or regex.
Question-first
#### Smart Lookup
Best when you want notes about an idea even if the note does not contain your search terms.
Note-first
#### Smart Connections
Best when you are already inside a note and want related material to update around your current context.
## How Smart Lookup works
Lookup is question-first. Your query becomes the reference point, and results are ranked by semantic similarity instead of exact word overlap.
### Ask naturally
Start with the idea, topic, or question you want to recover. Write it the way you would ask yourself.
### Read ranking as a signal
Similarity scores help rank the results inside one lookup. Expand a few candidates to confirm relevance before you act.
### Choose the right granularity
Use sources when you want broader note discovery. Use blocks when you want more precise sections without hunting inside long notes.
### Simple query formula
**Topic + context nouns + desired output**
**Stronger query:** "My notes about reducing information overload while researching - list the best steps."
**Fast refinement:** add 1-2 constraints such as project name, domain, people, tools, or output type.
Need the deeper walkthrough? Open the
[search guide](/smart-lookup/search/)
. Need to tune sources, blocks, limits, or filters? Use
[Connections settings](/smart-connections/settings/)
.
## From question to grounded action
Lookup is strongest when it hands off to the next workflow instead of ending at the results list.
[Find
### Smart Lookup
Recover the most relevant notes when you start from a question instead of an anchor note.](/smart-lookup/search/)
[Package
### Smart Context
Turn strong Lookup matches into reusable context so your next prompt starts grounded.](/smart-context/)
[Converse
### Smart Chat
Ask follow-up questions with the right notes already attached, then keep the thread tied to the work.](/smart-chat/)
Already inside a note and want related material to update around it automatically? Use
[the Connections view](/smart-connections/list-feature/)
for note-first discovery.
## What makes Smart Lookup trustworthy
- ✓
Lookup is additive, not a replacement for exact search. Use Obsidian Search for exact phrases and operators, and use Lookup for meaning-level discovery.
- ✓
Results stay inspectable. Expand items, confirm relevance, and then open, link, or export what deserves action.
- ✓
Lookup uses the same semantic layer as Smart Connections, so the mental model stays consistent across question-first and note-first workflows.
- ✓
When your vault grows, you can tune result type, limits, and filters in
[Connections settings](/smart-connections/settings/)
instead of guessing why the list changed.
## FAQ
Quick answers for people deciding whether Lookup is the right search surface.
What is Smart Lookup for Obsidian?
Smart Lookup is question-first semantic search for Obsidian. It lets you ask plain-language questions and surfaces notes by meaning, not just exact text overlap.
When should I use Smart Lookup instead of Obsidian Search?
Use Lookup when you remember the idea but not the exact phrase. Use Obsidian Search when you need exact phrases, tags, operators, file names, or regex.
How is Smart Lookup different from Smart Connections?
Smart Lookup is question-first: you start with a query. Smart Connections is note-first: you start from the note you are already viewing and let related results update around it.
Why did my exact phrase not appear near the top?
That can be normal in semantic search. Lookup ranks by meaning, not literal word overlap, so a note with different wording may outrank a note that repeats your exact phrase without matching your intent.
Should I return sources or blocks?
Start with sources when you want broader discovery. Switch to blocks when you keep opening long notes and hunting for the right section.
What should I do after I find the right notes?
Open the best match, link it into the working note, build a short reading trail, or export the strongest matches into
[Smart Context](/smart-context/)
for grounded AI work.
When you remember the idea but not the words, Smart Lookup turns recall friction into forward motion.
[See the Smart Lookup guide](/smart-lookup/search/)
[See Smart Connections](/smart-connections/)
---
## Smart Graph
canonical: https://smartconnections.app/smart-graph/
html_url: https://smartconnections.app/smart-graph/
markdown_url: https://smartconnections.app/smart-graph.md
llms_url: https://smartconnections.app/smart-graph/llms.txt
last_modified: 2026-09-07T19:38:08.082Z
usage_notes: |-
Use this page to answer questions about Smart Graph.
excerpt: |-
Semantic graph for Obsidian Smart Graph: see patterns across your notes Smart Graph arranges indexed notes by semantic similarity so you can explore regions, possible bridges, and outliers — then open the notes behind them. Get Smart Graph with Pro See the replays Available with Pro on Obsidian desktop. Choose a scope Whole vault Neighborhood Focus nodes Choose a replay Semantic cascade Graph…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Semantic graph for Obsidian
# Smart Graph: see patterns across your notes
Smart Graph arranges indexed notes by semantic similarity so you can explore regions, possible bridges, and outliers — then open the notes behind them.
[Get Smart Graph with Pro](/pro-plugins/)
[See the replays](#replays)
Available with Pro on Obsidian desktop.
Choose a scope
Whole vault
Neighborhood
Focus nodes
Choose a replay
Semantic cascade
Graph reveal
[Open the Whole vault Semantic cascade video](/assets/graph-replay-v2/whole-vault-cascade.mp4).
7.5 s loop · Silent · 1,050 sources in this recording
Play replay
Restart
[View larger ↗](/assets/graph-replay-v2/whole-vault-cascade.mp4)
**Whole vault · Semantic cascade** — The replay moves through representative notes and semantic regions, then returns to the whole-vault map.
These are separate recorded examples, not a continuous scope-changing workflow.
Use the video controls to play. All six edited replays: [Whole vault: Semantic cascade](/assets/graph-replay-v2/whole-vault-cascade.mp4) · [Whole vault: Graph reveal](/assets/graph-replay-v2/whole-vault-reveal.mp4) · [Neighborhood: Semantic cascade](/assets/graph-replay-v2/neighborhood-cascade.mp4) · [Neighborhood: Graph reveal](/assets/graph-replay-v2/neighborhood-reveal.mp4) · [Focus nodes: Semantic cascade](/assets/graph-replay-v2/focus-nodes-cascade.mp4) · [Focus nodes: Graph reveal](/assets/graph-replay-v2/focus-nodes-reveal.mp4).
Read the map
## Use the map to choose where to look next
Regions, possible bridges, and outliers can direct attention across a growing body of notes.
Region
A dense group can suggest a shared theme worth investigating.
[Understand regions](/docs/graph/#understand-cluster-regions)
Bridge
A generated bridge between regions can point to an overlap worth checking. It is not a link written in your notes.
[Understand bridges and links](/docs/graph/#understand-visible-bridges)
Outlier
A small or distant group can suggest a niche topic, a coverage gap, or an unexpected note worth checking.
[Interpret Graph patterns](/docs/graph/#interpret-clusters-representatives-and-affinities)
Why another graph
## Links show what you connected. Smart Graph adds a semantic view.
Obsidian native graph
### Trace explicit links
- Shows links already written into the vault.
- Shows how your notes are explicitly linked.
Smart Graph
### See notes that are close in meaning
- Groups indexed source notes by semantic similarity.
- Overlay authored links to compare semantic proximity with the links you created.
Use both when you want to compare semantic proximity with the links you created. See
[the native Graph comparison](/smart-graph/faq/#how-is-smart-graph-different-from-obsidians-native-graph)
and
[how Graph lines work](/docs/graph/#understand-visible-bridges).
Move through the map
## Move from the landscape to sources you can inspect
Search finds labels and paths, selection keeps chosen notes visible, and scope redraws the map around a smaller source set.
Search the graph
Find matching labels and paths while the rest stays visible.
[Use Graph search](/docs/graph/#search-the-graph)
Select notes to review
Keep selected note names visible in the Notes panel while you decide which sources to inspect or use.
[Use the Notes panel](/docs/graph/#use-the-notes-panel)
Focus the view
Render the graph around a cluster, selected notes, or a Neighborhood.
[Compare Graph scopes](/docs/graph/#choose-a-graph-scope)
Follow the trail
## Follow a pattern back to the notes
The graph directs attention; the original notes provide the evidence. Select a node or region, confirm its notes, and open them before deciding what the pattern means.
1. 1
Select one node or a small region.
2. 2
See the selected note names in the Notes panel.
3. 3
Open the notes you want to inspect in Obsidian.
Follow the first note-opening walkthrough in
[Getting Started](/smart-graph/getting-started/#first-win-open-one-related-note).
Smart Graph can open a selected source directly in Obsidian.
[How opening source notes works](/docs/graph/#smart-graph-actions-open).
Carry it forward
## Carry reviewed sources into the next step
- **Inspect sources:** Open selected notes in Obsidian.
- **Carry them forward:** Copy links, create a note or Canvas, or add selected notes to Smart Context when available.
- **Keep exploring:** Save the graph as an image or focus the map on a cluster, selected nodes, or a Neighborhood.
[Explore Smart Graph actions](/docs/graph/#use-smart-graph-actions)
Choose by question
## Different questions need different starting points
Smart Graph starts with a set of notes and asks what patterns appear. Other surfaces start from a current note, a typed question, exact text, or authored links.
### Smart Graph
Explore patterns across several notes.
[### Connections
Surface related notes from the note in view.](/smart-connections/)
[### Lookup
Search semantically from a typed phrase or question.](/smart-lookup/)
### Obsidian Search
Find exact text, paths, tags, operators, or regex.
### Native Graph
Explore authored links between notes.
How it works
## Semantic similarity shapes the layout
Smart Graph uses embeddings to place indexed notes closer together when their content is semantically similar. The selected scope determines which notes enter the map.
What can change
## The view changes with its inputs
What appears can change with the indexed sources, embedding model, scope, and layout settings.
[Learn how to interpret Graph](/docs/graph/#interpret-clusters-representatives-and-affinities)
Available with Pro
## Add a semantic view when scale makes relationships harder to see
Smart Graph is available through Pro on desktop. The sources you want to explore must be indexed; a current-note Neighborhood also needs the active note in that index. Check the live Smart Plugins Store for current access.
[Get Smart Graph with Pro](/pro-plugins/)
[Open Getting Started](/smart-graph/getting-started/)
Questions before you add it
## Smart Graph FAQ
What is Smart Graph for Obsidian?
+
Smart Graph arranges indexed notes by semantic similarity into regions you can explore, with source nodes you can select and open.
How is Smart Graph different from Obsidian's native graph?
+
Obsidian's native graph centers authored links. Smart Graph starts from semantic similarity and can also overlay authored links when **Show links** is active.
What do the lines mean?
+
Semantic bridges are generated lines between cluster representatives. **Show links** adds authored links with arrowheads that show direction; **Hide links** removes that overlay.
Are the nodes real notes I can open?
+
Yes. Nodes resolve to indexed source notes. Select a node to inspect or open its source in Obsidian.
[Learn about Graph nodes](/docs/graph/#understand-source-nodes)
Does Smart Graph help if my notes are not heavily linked?
+
Yes. Because Smart Graph uses semantic similarity, it can form Neighborhoods even when authored links are sparse.
What can I do after I review a region?
+
Inspect sources by opening selected notes. Carry a reviewed set forward by copying links, creating a note or Canvas, or adding it to Smart Context when available. Continue exploring by saving the graph image or changing scope.
What does Smart Graph require?
+
Smart Graph is a Pro desktop plugin. It needs eligible indexed Smart Sources and current embeddings. A current-note Neighborhood also requires the active note to be indexed.
Is Smart Graph experimental?
+
The current Store lists Smart Graph in the main Pro catalog rather than under Experimental. Check the live Store for the latest grouping.
## See where to look next
Get Smart Graph through Pro, start from a note or source set you understand, and choose what to inspect next.
[Get Smart Graph with Pro](/pro-plugins/)
[Open Getting Started](/smart-graph/getting-started/)
[Documentation](/docs/graph/)
[FAQ](/smart-graph/faq/)
---
## Smart Context
canonical: https://smartconnections.app/smart-context/
html_url: https://smartconnections.app/smart-context/
markdown_url: https://smartconnections.app/smart-context.md
llms_url: https://smartconnections.app/smart-context/llms.txt
last_modified: 2026-09-07T19:38:08.071Z
usage_notes: |-
Use this page to answer questions about Smart Context.
excerpt: |-
Smart Context Stop pasting vague prompts. Build grounded context packages from your actual notes, then reuse them across ChatGPT and other AI tools. Install Smart Context (Core, free) Explore Context Pro Smart Context runs inside Obsidian and exports to your clipboard. Content leaves when you paste it into another tool or send it through an enabled provider workflow. Copy folders | Follow links…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
# Smart Context
Stop pasting vague prompts.
Build grounded context packages from your actual notes, then reuse them across ChatGPT and other AI tools.
[Install Smart Context (Core, free)](https://obsidian.md/plugins?id=smart-context)
[Explore Context Pro](/pro-plugins/)
Smart Context runs inside Obsidian and exports to your clipboard. Content leaves when you paste it into another tool or send it through an enabled provider workflow.
Copy folders | Follow links by depth | Save named contexts | Keep prompts token-aware | Standardize output with templates | Export agent-ready bundles (Pro)
[First win](#quick-start)
[Workflows](#workflows)
[Core vs Pro](#core-vs-pro)
[Docs](#docs)
[FAQ](#faq)
Fastest first win: follow the
[Getting Started guide](/smart-context/getting-started/)
and copy one reviewable context bundle today.
## Copy your first reviewable context bundle
Better AI output usually starts before the model.
Smart Context helps you copy the right notes in the right shape so the assignment stays grounded and reviewable.
Start with the smallest context that makes the assignment understandable.
Fastest path
### Copy the current note at Depth 0 or 1
Best when one note already anchors the assignment. Use Depth 0 when the note is enough and Depth 1 only when direct links hold ground truth.
Learn:
[Copy the current note with link-depth control](/smart-context/clipboard/current/)
Selected scope
### Copy selected notes or a project folder
Best when the scope is already chosen. Copy a small folder, file-nav selection, or short set of notes instead of the whole vault.
Learn:
[Use file navigator actions to copy notes and folders as context](/smart-context/file-nav-actions/)
Repeatable work
### Build a reusable context pack
Best when you repeat the same work. Save a named context once, then reuse it across chats and projects.
Learn:
[Build and save reusable context packs](/smart-context/builder/)
Want outputs that stay consistent? Use
[templates in Settings](/smart-context/settings/)
so every export has the same structure (headers, metadata, file tree).
Want the shortest reliable loop? Copy a note at Depth 0 or 1, then prompt:
"Use only this context. Extract constraints first, then propose next actions."
### When should you use each Smart Context path?
- **Need first value fast?** Start with [current-note copy](/smart-context/clipboard/current/).
- **Need reusable packs?** Build named context sets in the [Builder](/smart-context/builder/).
- **Need deterministic output format?** Configure template wrappers in [Settings](/smart-context/settings/).
- **Need external files for coding workflows?** Pin manifests with the [Codeblock workflow](/smart-context/codeblock/) (Pro).
## Is Smart Context the right workflow for you?
Use Smart Context when your notes already contain the answer, but your AI replies still drift because the right evidence is not packaged together.
### Great fit
- You already keep project notes, meeting notes, or research hubs in Obsidian.
- You need consistent context bundles for repeated tasks (status updates, prep briefs, drafting).
- You want to review exactly what travels before sending context to any model.
### Start here if you are new
- Run the [Getting Started guide](/smart-context/getting-started/) first.
- Then [copy the current note](/smart-context/clipboard/current/) and select a link depth.
- When you need repeatable packs, build a [named context set](/smart-context/builder/).
## Pick a workflow
This is where Smart Context pays off: when your notes are right, but your prompts are not.
Choose the job that matches your current friction and copy a pack that fixes it.
### Project working set
If every new chat starts with backstory, save a named pack once and reuse it.
Include: brief, constraints, decisions, latest status. Build it in the
[Builder](/smart-context/builder/).
Prompt starter
"Using only this context, summarize current state, risks, and the top 5 next actions. Cite the note title for each claim."
### Meeting continuity
If decisions keep getting lost between meetings, export a reliable prep + follow-up bundle.
Include: agenda, last notes, decisions, open tasks. Copy via
[Clipboard](/smart-context/clipboard/)
(or save as a named pack).
Prompt starter
"Draft a prep brief and follow-ups with owners. Pull action items from the context and keep wording concise."
### Research brief
If your hub note cites the real sources, copy it with Depth 1 so the references come along.
Include: hub note + cited notes. Use
[Depth 1](/smart-context/clipboard/current/)
to stay grounded.
Prompt starter
"Summarize, list open questions, and cite sources by note title. If evidence conflicts, call it out explicitly."
### Voice + draft pack
If your writing help keeps drifting, bundle your rules and examples with the draft.
Include: outline, draft, tone rules, 1-3 examples. Standardize output with
[templates](/smart-context/settings/).
Prompt starter
"Continue the draft in my voice. Follow the outline. Use only the provided references for claims."
### Context budget triage
If context keeps exceeding limits, treat depth and blocks as budget controls.
Start at Depth 0. Expand only when needed. Use the size estimate as a budget signal.
Prompt starter
"If this is too large, tell me what to remove first (in order) while preserving constraints and ground truth."
### Visual troubleshooting (Context Pro)
If the truth lives in screenshots, diagrams, or PDFs, export visuals alongside text.
Use
[Copy context with images and PDFs](/smart-context/clipboard/media/)
and paste into an image-capable model.
Prompt starter
"Use both the text and images. Call out any visual gaps or mismatches with the written notes."
Coders and agents: pin external repos and files inside a note with the
[Codeblock workflow](/smart-context/codeblock/)
(Context Pro).
## Smart Context Core vs Context Pro
Start with Core for trusted, note-native context. Add Context Pro when the assignment needs media, external files, repos, Bases, or advanced controls.
Core plugin
### Smart Context Core
Build precise bundles of notes and reuse them across prompts and projects.
- ✓
Copy current note, selected notes, or folders with link depth control.
- ✓
Save named contexts so your best bundles are always one click away.
- ✓
Template-driven exports for stable formatting and metadata.
Pro plugin
Advanced context sources
### Context Pro
Advanced tools for context engineering: external sources, media-aware exports, and agent-ready outputs.
- ✓
Export context with images and PDFs so visual ground truth is not lost.
- ✓
Include external repos/files when implementation truth lives outside the vault.
- ✓
Use advanced sources only when the assignment needs them.
Already a monthly, yearly, or Founding supporter? Your supporter perks already include Context Pro features. No extra Pro plugins subscription is required.
## Docs and next steps
Start simple, then level up to reusable packs, predictable templates, and Pro workflows.
Get started
Getting Started
Screenshots + the fastest first-win workflows.
[Copy loop
Copy notes and folders as AI-ready context
Copy current note, selected notes, or folders with depth + size estimates.](/smart-context/clipboard/)
[Reuse
Build and save reusable context packs
Save trusted context packs and reopen them instantly.](/smart-context/builder/)
[Format control
Control Smart Context templates and export format
Standardize wrappers, metadata, and file tree output.](/smart-context/settings/)
[Context Pro
Attach a Smart Context codeblock to a note
Pin external repos/files in notes for agent-ready context.](/smart-context/codeblock/)
[File navigator
Use file navigator actions to copy notes and folders as context
Copy selected notes, folders, or curated file-nav selections without opening another surface.](/smart-context/file-nav-actions/)
[Current note
Copy the current note with link-depth control
Start from the note you are already working in and choose how far links should expand.](/smart-context/clipboard/current/)
[Canvas
Copy context from Obsidian Canvas files
Use a visual project map as the starting point for copied AI context.](/smart-context/canvas/)
[Context Pro
Copy context with images and PDFs
Include visual ground truth when text alone is not enough.](/smart-context/clipboard/media/)
[Saved packs
Reopen and reuse saved named contexts
Turn recurring project context into a reusable working set.](/smart-context/builder/named/)
[Structure
Apply reusable prompt structure with Smart Templates
Context supplies the evidence; Templates supplies the repeatable form.](/smart-templates/)
[Upgrade
Unlock advanced Smart Plugins Pro workflows
Context Pro, Smart Chat, and more advanced workflows.](/pro-plugins/)
## FAQ
Quick answers to the highest-intent pre-install questions about compatibility, privacy, first wins, token budgets, and Pro workflows.
What is Smart Context for Obsidian?
Smart Context helps you build grounded context packages from your actual Obsidian notes, then reuse them across ChatGPT and other AI tools. It reduces vague prompts by helping you copy the right notes in a clean, repeatable format.
Can I use Smart Context with ChatGPT, Claude, Gemini, or Smart Chat?
Yes. Smart Context is clipboard-based, so you can paste exported context into ChatGPT, Claude, Gemini, [Smart Chat](/smart-chat/), or another AI tool. The same context bundle can be reused anywhere that accepts pasted text.
Does Smart Context keep my notes private and local-first?
Smart Context runs inside Obsidian and exports to your clipboard. Content leaves when you paste it into another tool or send it through an enabled provider workflow. Review providers and data boundaries in [Smart Environment settings](/smart-environment/settings/) and the [Privacy Policy](/legal/privacy-policy/).
What is the fastest way to get started with Smart Context?
Copy the current note with Depth 0 or 1, then paste it into the AI tool or Smart Chat thread you already use. You know it worked when the copied bundle is small enough to inspect and includes the context the assignment needs.
Should I copy the current note, a whole folder, or build a reusable context pack?
Use the current note when one note anchors the task, a folder when the project already lives together, and a reusable pack when you need a curated working set you will reopen across sessions.
How does link depth work in Smart Context?
Depth 0 copies the starting note. Depth 1 adds directly linked notes. Depth 2 adds links of links, and deeper levels expand further. Start small and expand only when the extra evidence is needed. See [Clipboard + depth](/smart-context/clipboard/).
How do I keep Obsidian AI context under token limits?
Start at Depth 0, expand only when needed, and use the size estimate as a budget signal. If one note is too large, switch to a smaller selection or a curated pack instead of copying the entire folder.
Can I save reusable context packs or named contexts?
Yes. The [Builder](/smart-context/builder/) lets you create, name, reopen, and reuse trusted context sets across chats and projects so you do not have to rebuild the same bundle every time.
Can I customize the exported prompt format and file tree?
Yes. [Settings](/smart-context/settings/) lets you customize context templates, item templates, wrappers, file paths, and file tree output so exports stay stable across workflows.
What does Context Pro add over the free Smart Context Core plugin?
Core covers fast selection, link depth, named contexts, and template-driven exports. [Context Pro](/pro-plugins/) adds media-aware copy flows, external files and repos, and deeper agent-ready workflows when the assignment needs them.
Can I include images, PDFs, repos, or external files in my AI context?
Yes, with Context Pro. [Copy with media](/smart-context/clipboard/media/) handles images and PDFs, and the [Codeblock workflow](/smart-context/codeblock/) lets you include external repos or files alongside note-based context.
Can I use Smart Connections results to build a context pack?
Yes. You can send results from the Connections view to Smart Context, remove noise, and copy the curated pack. See the [Connections view guide](/smart-connections/list-feature/).
Build your first context pack now and watch your prompts level up. Start with the free Core plugin, then add Context Pro when you need media, external sources,
and agent-ready bundles.
[Install Smart Context (Core, free)](https://obsidian.md/plugins?id=smart-context)
[See Getting Started](/smart-context/getting-started/)
---
## Smart Environment
canonical: https://smartconnections.app/smart-environment/
html_url: https://smartconnections.app/smart-environment/
markdown_url: https://smartconnections.app/smart-environment.md
llms_url: https://smartconnections.app/smart-environment/llms.txt
last_modified: 2026-09-07T19:38:08.071Z
usage_notes: |-
Use this page to answer questions about Smart Environment.
excerpt: |-
Smart Environment Smart Environment is the local-first runtime that unifies Smart Plugins, adapters, and automations inside your vault without copying notes to remote servers. It orchestrates context, schedules actions, and exposes a consistent interface so every new capability plugs in without adding another siloed app. Unified runtime. One orchestrator coordinates Smart Chat, Smart Connections,…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
# Smart Environment
Smart Environment is the local-first runtime that unifies Smart Plugins, adapters, and automations inside your vault without copying notes to remote servers. It orchestrates context, schedules actions, and exposes a consistent interface so every new capability plugs in without adding another siloed app.
- **Unified runtime.** One orchestrator coordinates Smart Chat, Smart Connections, and other Smart Tools so they share context and resources.
- **Local-first architecture.** Your notes stay on your device by default, with explicit opt-ins for remote AI services.
- **Modular plugins.** Each Smart Plugin focuses on a specific capability, reusing shared components for context management and data access.
- **Adapter layer.** Standard adapters translate between local models, remote APIs, and vault data, avoiding brittle single-purpose integrations.
- **Privacy guardrails.** Actions run against files you control with explicit opt-ins before anything leaves your machine.
- **Observable workflows.** Structured logs make it easy to review, reproduce, and tune automations.
- **Cognitive architecture.** System design inspired by human cognition improves adaptability and extensibility.
New here? Start with [Smart Milestones](https://smartconnections.app/smart-environment/milestones/) to unlock the fastest first wins across Smart Plugins.
## Why Smart Environment matters
- Provides a consistent backbone so new plugins ship faster without reinventing pipelines.
- Keeps sensitive knowledge bases private and auditable.
---
## Smart Chat
canonical: https://smartconnections.app/smart-chat/
html_url: https://smartconnections.app/smart-chat/
markdown_url: https://smartconnections.app/smart-chat.md
llms_url: https://smartconnections.app/smart-chat/llms.txt
last_modified: 2026-09-07T19:38:08.055Z
usage_notes: |-
Use this page to answer questions about Smart Chat.
excerpt: |-
Smart Chat Keep AI conversations where the work lives. Start with Core codeblocks to save provider thread links in the notes they serve, then use Smart Chat API Extension when you need model choice, reviewed sources, and API Extension thread control inside Obsidian. Install Smart Chat Core Use Smart Chat API Extension Core: saved provider thread links in Markdown | API Extension: model, sources,…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
# Smart Chat
Keep AI conversations where the work lives. Start with Core codeblocks to save provider thread links in the notes they serve, then use Smart Chat API Extension when you need model choice, reviewed sources, and API Extension thread control inside Obsidian.
[Install Smart Chat Core](https://obsidian.md/plugins?id=smart-chatgpt)
[Use Smart Chat API Extension](/smart-chat/api/)
Core: saved provider thread links in Markdown | API Extension: model, sources, thread, and response visible before trust
[Core first](#core-first)
[API Extension](#api-extension)
[Which path?](#core-vs-api)
[Guides](#docs)
[FAQ](#faq)
## Choose by the job
Core is complete for provider-thread continuity. Smart Chat API Extension adds model and source control when the conversation needs it.
Use Core when...
### The thread should stay attached to a note
Save official provider thread links in Markdown, return later from the originating note, and track active or done state.
Use API Extension when...
### You need to set up API Models
Choose a configured local or cloud model, attach or retrieve sources, review them before sending, and manage API Extension threads.
## Start with Core: save one thread link
Smart Chat Core keeps provider conversations attached to the notes they serve.
1. Step 1
### Open the note
Use the project, meeting, research, decision, or draft note where the AI thread belongs.
2. Step 2
### Add a provider codeblock
Start or open a provider thread using the web UI you already use, such as ChatGPT, Claude, Gemini, Grok, or Perplexity.
3. Step 3
### Save the thread link
The note now contains a thread link or active/done state that can be reopened later.
You know it worked when the note contains a provider thread link you can reopen later.
[Learn Core codeblocks](/smart-chat/codeblock/)
## Advanced configuration: Add Smart Chat API Extension
API Extension is the Smart Chat path for local or cloud models you have set up, source review before sending, custom instructions, and API Extension thread recovery.
1. Proof 1
### Confirm model
See the selected/default model before sending.
2. Proof 2
### Attach source
Add one known source or use Lookup to retrieve candidates for review.
3. Proof 3
### Use thread
Ask one outcome-focused question in the current API Extension thread.
4. Proof 4
### Review response
Promote only the parts that match your outcome and sources.
You know it worked when the model, sources, thread, and response are visible before you trust the answer.
[Start API Extension workflow](/smart-chat/api/)
## Smart Chat guides
Pick the guide for the thread state you need to manage.
Start here
Getting Started
Add one Core codeblock and save one provider thread link.
[Core
Smart Chat codeblocks
Save provider thread links and active/done state in Markdown.](/smart-chat/codeblock/)
[API Extension
Model and source control
Confirm model, review sources, ask one outcome-focused question.](/smart-chat/api/)
[Core dashboard
Chat Inbox with Dataview
List note-attached `chat-active` and `chat-done` thread links.](/smart-chat/thread-dashboard/)
[API Extension dashboard
Thread Manager
Open, rename, search/filter, and clean up API Extension thread records.](/smart-chat/thread-manager/)
[Mobile and sync
Mobile/sync FAQ
Separate Core Markdown link sync from API Extension `.smart-env` storage.](/smart-chat/mobile-sync/)
## The thread is not the deliverable
Smart Chat helps preserve or control the conversation. Your note remains the trusted place where reviewed decisions, context, next actions, and deliverables belong.
1. Open the note that owns the work.
2. Reopen the saved provider thread or API Extension thread.
3. Compare the output against the outcome and sources.
4. Promote only the useful parts.
5. Mark the thread done or keep it active with a clearer next action.
## See Smart Chat in action
Watch a thread stay attached to the note it serves.
## FAQ
Do I need an API key?
Not for Core chat codeblocks. Core uses provider web UIs and saves provider thread links in Markdown. Smart Chat API Extension uses local or cloud API models you have configured in Smart Environment.
What does Core store?
Core stores provider thread links and active/done state in your notes. It does not store the full provider transcript by default.
What does Smart Chat API Extension add?
It adds the API integration for configured local or cloud models, reviewed sources before sending, custom instructions, include/exclude controls, and API Extension thread history/management.
Does this work on mobile?
Core codeblock links sync as Markdown and can be opened from mobile. The full embedded provider web UI is desktop-first. Smart Chat API Extension stores thread data in `.smart-env`, which is not recommended to sync by default. See the mobile/sync FAQ for setup guidance.
Is Thread Manager the same as a Chat Inbox?
No. Chat Inbox reads Core `chat-active` and `chat-done` Markdown fields. Thread Manager manages Smart Chat API Extension thread records.
Start by saving one provider thread link in the note it serves. Add Smart Chat API Extension only when choosing the model and reviewing sources before sending matters.
[Install Smart Chat Core](https://obsidian.md/plugins?id=smart-chatgpt)
[Start API Extension guide](/smart-chat/api/)
---
## Smart Connections
canonical: https://smartconnections.app/smart-connections/
html_url: https://smartconnections.app/smart-connections/
markdown_url: https://smartconnections.app/smart-connections.md
llms_url: https://smartconnections.app/smart-connections/llms.txt
last_modified: 2026-09-07T19:38:08.055Z
usage_notes: |-
Use this page to answer questions about Smart Connections.
excerpt: |-
Related notes for Obsidian Smart Connections: related notes while you work Smart Connections uses the note in front of you to surface related notes and passages, including ideas you did not think to search for or link. Earlier research and decisions can contribute again while you work. Install Smart Connections free See an example Free Core plugin No API key required Local embeddings by default…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Related notes for Obsidian
# Smart Connections: related notes while you work
Smart Connections uses the note in front of you to surface related notes and passages, including ideas you did not think to search for or link. Earlier research and decisions can contribute again while you work.
[Install Smart Connections free](https://community.obsidian.md/plugins/smart-connections)
[See an example](#proof)
- Free Core plugin
- No API key required
- Local embeddings by default
Open a result to read the matching passage without leaving the note you are working in.
Why it matters
## A growing vault can outpace what you can keep in mind
Research, conversations, decisions, and AI-created notes can accumulate faster than you can integrate them. Work gets repeated, decisions reopen without their reasoning, and useful ideas fail to reach the next task.
### Avoid repeated research
Bring earlier interviews, evidence, and comparisons back before you repeat the work.
### Carry decisions forward
Recover the constraint or rationale before the same debate starts again.
### Find the idea, not the filename
Recover related thinking when the title, folder, or exact wording is no longer in memory.
Choose the starting point
## Each tool starts from a different clue
S
Exact search
Start with words, filenames, headings, tags, syntax, or regex you remember.
B
Backlinks
Start with relationships already written into your notes.
C
Smart Connections
Start with the note in view and surface notes with similar meaning, even without matching wording or links.
[Compare Connections with exact search](/smart-connections/faq/#how-is-this-different-from-keyword-search)
A first useful result
## Start with one note you already understand
1. 1
Open a note with enough text to show what you are working on.
2. 2
Scan the first few results and expand one promising match.
3. 3
Open the original note, bring what applies into the work, and leave the rest behind.
Follow the full first-use path in
[Getting Started](/smart-connections/getting-started/#find-one-useful-related-note).
[Learn about fixed-target behavior](/docs/connections/#auto-refresh-and-fixed-target-behavior).
Open a result while Connections keeps comparing against the note you started from.
Where it helps
## Let past work contribute to new work
Writing
### Bring earlier evidence into the draft
Recover interview language, examples, and earlier framing while you write.
Projects
### Resume without reopening every decision
Bring constraints, experiments, and discarded options back into view when work resumes.
Meetings
### Prepare with the surrounding context
See related people, plans, and prior conversations before the meeting starts.
Core and Pro
## Start with Core. Add Pro for finer placement and ranking.
Core puts related notes in the Connections view and at the end of notes. Pro adds passage-level suggestions, relevance-ranked Base rows, and advanced filtering and ranking controls.
In either Core surface, choose a list for quick scanning or add the Connections mini graph to see the current result set visually.
Core
### Connections view
Keep a dedicated related-note list beside the note in view, then expand or open results without leaving the workspace.
[Explore the Connections view](/docs/connections/#open-the-connections-view)
Core
### Footer connections
Put related notes at the end of each note, which is useful on mobile or when the sidebar is closed.
[Explore Footer connections](/docs/connections/#show-related-notes-at-the-bottom-of-an-obsidian-note)
Pro
### Inline connections
See related material for the paragraph, heading, or block you are editing without leaving the editor.
[Explore Inline connections](/docs/connections/#find-related-notes-beside-the-text-you-are-writing)
Pro
### Connections in Bases
Use a reference note to rank Base rows by semantic relevance.
[Explore Connections in Bases](/docs/connections/#rank-obsidian-bases-rows-by-semantic-relevance)
Local by default
## Start without an API key
Smart Connections Core runs its standard related-note workflow with local embeddings. Cloud providers are involved only when you choose to configure them.
Useful offline after indexing
Local retrieval can keep working after indexing. Installation, updates, model downloads, and cloud workflows still need internet.
You decide what to use
Connections suggests related notes. You choose what to open, link, reuse, or ignore.
See what is indexed
Smart Environment shows included sources, exclusions, embedding status, and index readiness.
[API key FAQ](/smart-connections/faq/#do-i-need-an-api-key)
[Smart Environment settings](/smart-environment/settings/)
[Privacy Policy](/legal/privacy-policy/)
Questions before you install
## Smart Connections FAQ
Do I need an API key?
+
No. Smart Connections Core uses local embeddings by default, so its standard related-note retrieval does not need an API key. An API key is only needed for optional cloud providers or integrations you choose to configure.
[Read the full API key answer](/smart-connections/faq/#do-i-need-an-api-key)
Are Footer connections included in Core?
+
Footer connections are included in Core, including both **List only** and **Version 4.0 (Graph + List)**.
[Read the Footer edition answer](/smart-connections/faq/#is-footer-connections-core-or-pro)
Is the Connections mini graph the same as Smart Graph?
+
No. The Connections mini graph is a Core display for the current result list. Smart Graph is the separate Pro desktop plugin for exploring semantic structure across larger source sets.
How is Smart Connections different from keyword search?
+
Exact search starts with words or syntax you remember. Smart Connections starts with the note in view and surfaces related notes or passages even when the wording or title differs.
What does Connections Pro add?
+
Pro adds Inline connections, Connections in Bases, and advanced filtering, scoring, and ranking controls. The Connections view and Footer connections are included in Core.
Does Smart Connections send my notes to the cloud?
+
Core retrieval uses local embeddings by default. A cloud embedding provider receives eligible source text only when you configure one. A cloud chat provider receives the prompt and context for the request you send.
Can I use Smart Connections on mobile?
+
Yes. The Connections view and Footer connections are available on mobile; some controls vary by platform.
[Review Smart Plugins on mobile](/smart-plugins/mobile/)
## Bring past work back into the moment
Open the work already in front of you and see what relevant past notes can contribute.
[Install Smart Connections free](https://community.obsidian.md/plugins/smart-connections)
[Open Getting Started](/smart-connections/getting-started/)
[Documentation](/docs/connections/)
[FAQ](/smart-connections/faq/)
---
## Core Plugins
canonical: https://smartconnections.app/core-plugins/
html_url: https://smartconnections.app/core-plugins/
markdown_url: https://smartconnections.app/core-plugins.md
llms_url: https://smartconnections.app/core-plugins/llms.txt
last_modified: 2026-09-07T19:38:08.040Z
usage_notes: |-
Use this page to answer questions about Core Plugins.
excerpt: |-
Start here with Smart Plugins Start with core Smart Plugins for Obsidian Start with Smart Connections and get one useful result from your own notes. Add Smart Context, Smart Chat, and Smart Templates when the next workflow needs context, continuity, or structure. Need a side-by-side Copilot alternative comparison first? Review the migration path and feature tradeoffs. Core retrieval can run in…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Faq
url: https://smartconnections.app/faq/
- title: File Over App
url: https://smartconnections.app/file-over-app/
Start here with Smart Plugins
# Start with core Smart Plugins for Obsidian
Start with Smart Connections and get one useful result from your own notes. Add Smart Context, Smart Chat, and Smart Templates when the next workflow needs context, continuity, or structure.
Need a side-by-side [Copilot alternative](/obsidian-copilot/) comparison first? Review the migration path and feature tradeoffs.
Core retrieval can run in Obsidian, templates stay in Markdown, and provider workflows use the context you choose to send.
## Which Core plugin does what?
Use this quick map to decide where to start. Each plugin covers a different workflow layer, and Core remains useful before Pro.
| Plugin | What it is best for | Typical result |
| --- | --- | --- |
| Smart Connections | Finding related notes while you write | One useful related note from your own vault |
| Smart Context | Copying reviewable context from notes and folders | A smaller, clearer context bundle |
| Smart Chat | Keeping AI threads attached to the notes they serve | A thread that can be resumed from the originating note |
| Smart Templates | Turning current work and Markdown templates into structured prompts | A copied prompt with context, task, and trusted form |
## Core Smart Plugins for your Obsidian vault
Install Smart Connections first when you want to see relevant content. Add Context, Chat, and Templates when discovery needs packaging, the thread should stay with the note, or the output form repeats.
Core plugin
### Smart Connections
Related notes surface while you write.
- -
Open one real note and see what connects.
- -
Preview, drag, copy, or send strong matches to Smart Context.
- -
Use Lookup when the question, not the current note, is the anchor.
[Open in Obsidian marketplace](https://community.obsidian.md/plugins/smart-connections)
[Learn the Smart Connections workflow](/smart-connections/)
Core plugin
### Smart Context
Copy the context your AI work needs.
- -
Start with the current note or a small selected set.
- -
Choose Depth 0 or 1 so the bundle stays reviewable.
- -
Save named contexts when the same work repeats.
[Open in Obsidian marketplace](https://obsidian.md/plugins?id=smart-context)
[Learn the Smart Context workflow](/smart-context/)
Core plugin
### Smart Chat
Keep AI threads attached to the notes they serve.
- -
Store provider thread links inside the note where the work lives.
- -
Return later without hunting through browser tabs.
- -
Review output before it earns a place in the trusted note.
[Open in Obsidian marketplace](https://obsidian.md/plugins?id=smart-chatgpt)
[Learn the Smart Chat workflow](/smart-chat/)
Core plugin
### Smart Templates
Turn current work and Markdown templates into structured prompts.
- -
Start from the current note or selection instead of a blank prompt.
- -
Use context first, template second, output third.
- -
Copy a ready-to-run prompt for the AI tool you already use.
[Open in Obsidian marketplace](https://obsidian.md/plugins?id=smart-templates)
[Learn the Smart Templates workflow](/smart-templates/)
## How to install core plugins
Install the Core plugins from Obsidian's Community plugins tab. They share Smart Environment setup where relevant.
1. 1
Open Obsidian, go to Settings -> Community plugins, and turn on community plugins.
2. 2
Click Browse, then search for Smart Connections, Smart Context, Smart Chat, or Smart Templates.
3. 3
Install and enable the plugin that matches your current workflow. Start with one visible result before adding the next plugin.
Ready for more advanced workflows later? Smart Plugins Pro adds advanced controls for richer context, graph, generation, cleanup, model routing, performance, and approved vault actions.
[Explore Smart Plugins Pro](/pro-plugins/)
---
## Index
canonical: https://smartconnections.app
html_url: https://smartconnections.app
markdown_url: https://smartconnections.app/index.md
llms_url: https://smartconnections.app/index/llms.txt
last_modified: 2026-09-07T19:38:08.040Z
usage_notes: |-
Use this page to answer questions about Index.
excerpt: |-
Smart Plugins for Obsidian Find related notes while you work. Smart Connections shows related notes and passages beside the note you're working on, without a search query or existing links. Bring earlier research and decisions into your current task. Open in Obsidian View plugin listing Local embeddings by default No API key required Current note Related passage Enlarge screenshot ↗ Your…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Smart Plugins for Obsidian
# Find related notes
while you work. {#home-heading}
Smart Connections shows related notes and passages beside the note you're working on, without a search query or existing links. Bring earlier research and decisions into your current task.
Open in Obsidian
[View plugin listing](https://community.obsidian.md/plugins/smart-connections)
Local embeddings by default
No API key required
Current note
Related passage
Enlarge screenshot ↗
## Your vault can grow faster than you can make sense of it. {#problem-heading}
Research, decisions, and AI drafts accumulate. Useful background can get left out of the next task.
### Ideas you meant to revisit
You save an idea for later, then miss it when the project starts.
### Decisions without the background
A decision is recorded, but the research behind it is hard to find.
### AI drafts without context
Drafts pile up, separate from the sources and goals they were meant to support.
## Bring your notes into the next AI task. {#ecosystem-heading}
Smart Connections works on its own. Add other plugins only as needed.
01
### Find related notes
[Smart Connections](/smart-connections/) starts from your current note. [Smart Lookup](/smart-lookup/) starts from a question.
02
### Choose what to include
[Smart Context](/smart-context/) lets you choose and review the sources for your AI request.
Enlarge screenshot ↗
03
### Keep AI work attached to the note
[Smart Chat Core](/smart-chat/) saves a link to the AI conversation in your note, so you can return to it as the work continues.
A conversation link stays with your note
Concept illustration: an Obsidian note contains a saved link that leads back to a conversation at the AI provider. This is not the Smart Chat interface.
Your note
Saved link
AI conversation
at the provider
Illustration: your note keeps the link, not a copy of the conversation.
[Explore all Smart Plugins →](/smart-plugins/)
Smart Graph · Pro on Obsidian desktop
## Explore your notes by meaning. {#graph-heading}
Smart Graph groups indexed notes by semantic similarity, whether or not you've linked them. See the patterns, then explore the notes behind them.
[Explore Smart Graph →](/smart-graph/#replays)
[Open the Whole vault Semantic cascade video](/assets/graph-replay-v2/whole-vault-cascade.mp4).
7.5 s loop · Silent · 1,050 sources in this recording
Play replay
Restart
[View larger ↗](/assets/graph-replay-v2/whole-vault-cascade.mp4)
**Whole vault · Semantic cascade** — The replay moves through representative notes and semantic regions, then returns to the whole-vault map.
Core and Pro
## Start free. Add the capabilities you need. {#editions-heading}
Smart Plugins Core
### Find and reuse your notes.
Use the Connections view beside your note, or Footer connections at the end.
Other free Smart Plugins help you review context and keep AI conversation links in your notes.
[Explore Smart Plugins Core →](/core-plugins/)
Smart Plugins Pro
### Visual exploration and finer controls.
Explore notes visually with Smart Graph or see passage-level suggestions with Inline connections.
Advanced ranking controls let you adjust how results are ordered.
[Explore Pro workflows →](/pro-plugins/)
## Let past work contribute again. {#start-heading}
Open the note you're working on. See which earlier ideas are worth bringing into your next draft or decision.
Smart Connections Core matches notes locally by default. Suggestions do not edit your notes.
[Open in Obsidian](obsidian://show-plugin?id=smart-connections)
[View plugin listing](https://community.obsidian.md/plugins/smart-connections)
[Open Getting Started →](/smart-connections/getting-started/)
## Smart Connections screenshot {#connections-proof-dialog-title}
Close
[Open original image →](/assets/connections-current-main-expanded-v4-8-1-documentation-1280x720-desktop-2026-08-07.png)
## Smart Context screenshot {#context-proof-dialog-title}
Close
[Open original image →](/assets/connections-to-context-current-v4-8-1-v3-4-1-documentation-960x540-desktop-2026-08-07.png)
---
## Pro Plugins
canonical: https://smartconnections.app/pro-plugins/
html_url: https://smartconnections.app/pro-plugins/
markdown_url: https://smartconnections.app/pro-plugins.md
llms_url: https://smartconnections.app/pro-plugins/llms.txt
last_modified: 2026-09-07T19:38:08.040Z
usage_notes: |-
Use this page to answer questions about Pro Plugins.
excerpt: |-
Smart Plugins Pro Pro plugins for Obsidian from the maker of Smart Connections Connections Pro, Context Pro, Chat Pro, Connect Pro, and large-vault performance in one 14-day free trial. Start with the workflow that already matters most in your vault: faster discovery in 1000+ note vaults, in-flow connections, reusable context, organized chat, or vault actions through Connect Pro. Start 14-day…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
Smart Plugins Pro
# Pro plugins for Obsidian from the maker of Smart Connections
## Connections Pro, Context Pro, Chat Pro, Connect Pro, and large-vault performance in one 14-day free trial.
Start with the workflow that already matters most in your vault: faster discovery in 1000+ note vaults, in-flow connections, reusable context, organized chat, or vault actions through Connect Pro.
[Start 14-day free trial](#pro-checkout-email)
[View pricing](#plans)
No charge today. Cancel before trial renewal to avoid charges.
Connections Pro
### Stay in flow while writing in large vaults
See relationships where you draft, tune the signal, and keep discovery responsive as your vault grows.
- Granular inline connections in the editor
- Faster Connections and Lookup in 1000+ note vaults with a local performance index
- Better signal in larger vaults via ranking and filtering controls
- Customizable connections algorithms and filters
Who this is for: writers, researchers, and Obsidian power users with large vaults.
Stop breaking flow as your vault grows.
Smart Graph
### See vault relationships as clusters and hubs
Explore how notes and blocks relate, then move from map-level insight back into writing, lookup, or cleanup.
- Visualize semantic clusters, hubs, and nearby notes
- Use the local performance index for calculation-heavy vault maps
- Explore result clusters before tuning Connections, Lookup, or Dedupe
- Built for large-vault relationship discovery
Who this is for: visual thinkers, researchers, and power users mapping large vaults.
Stop navigating your vault blind.
Context Pro
### Build reusable context packs for real AI work
Create grounded context once, then reuse it across prompts and agents.
- Named context packs with token visibility
- Add repositories and local files as context
- Include PDFs and images in exports for grounded outputs
Who this is for: developers, analysts, and researchers shipping with AI.
Stop rebuilding context.
Chat Pro
### A real chat workspace, not lost tabs
Keep long-running conversations organized inside Obsidian.
- API integration with chat models
- Vault-aware context lookup and review before sending
- Persistent searchable threads and per-thread instructions
Who this is for: builders who return to the same projects daily.
Stop losing threads.
Connect Pro
### Turn chat into vault actions
Stay in flow inside Obsidian when ChatGPT needs to create, append, triage, or query.
- Secure tunnel from Obsidian Desktop to the Official GPT
- Create notes, update daily notes, triage tasks, and query Bases
- Connect or disconnect anytime from the plugin
Who this is for: operators and builders who want chat to do real vault work.
Your vault becomes the control surface.
Pricing
## Choose your Smart Connections Pro trial
Start with 14 days free. All-access includes the Pro plugins available today plus large-vault performance.
Yearly saves ~17%
· 2 months free
Monthly for flexibility
switch to yearly to save $61
Monthly
Flexible billing, cancel anytime
Yearly
2 months free
Best value
Yearly selected. Save about 17% (around $61 per year) compared to paying monthly.
Monthly selected. Switch to yearly to get 2 months free (about 17% off).
Start here | Free
### Core Smart Plugins
$0 forever
- ✔
Essential product-specific features.
- ✔
Local-first defaults.
- ✔
Great for proving the workflow before you add more.
[Install core plugins (free)](https://smartconnections.app/core-plugins/)
Leaving soon
Best value · 2 months free
### Smart Connections Pro All-access
Advanced workflows and large-vault performance across the Smart Plugins suite. No charge today. Cancel before trial renewal to avoid charges.
$30/month
$299/year
Flexible billing, cancel anytime. Or switch to $299/year (about $25/month) to save $61 per year.
~ $25/month · Save $61 vs paying monthly.
- ✔
Access to all Pro plugins while your plan is active.
- ✔
New Pro plugins included automatically.
- ✔
What you get today: Connections Pro, Context Pro, Chat Pro, and Connect Pro.
- ✔
Large-vault performance index for Connections, Lookup, Smart Graph, inline connections, and Dedupe.
- ✔
Connect Pro turns chat into vault actions through the Official GPT and Obsidian CLI.
Connect Pro is included in All-access and currently requires Obsidian CLI on Obsidian 1.12+.
[Start monthly trial](#pro-checkout-email)
[Start yearly trial](#pro-checkout-email)
No charge today. Cancel anytime in 2 clicks.
1. 1) Start your 14-day free trial.
2. 2) Install or enable your Pro plugins.
3. 3) Use your vault workflows with full Pro access.
Roadmap pricing (coming soon): single-plugin and 3-plugin bundle options
Those options are in active roadmap planning and are not available for purchase yet.
If you want an update when they launch, join the waitlist:
[Join pricing waitlist](https://docs.google.com/forms/d/e/1FAIpQLSfzjnNFKKp1Z1_odShZtTRD84P8pFqrIemEG4p92MbC-nnyHQ/viewform)
Power-user performance
### Keep large-vault semantic workflows responsive
Pro is built for Obsidian power users whose vaults have grown beyond simple lookup: faster large-vault retrieval, deeper controls, inline connections, Smart Graph, and duplicate detection without giving up local-first architecture.
Connections, Lookup, inline connections, Smart Graph, and Dedupe reuse a faster local performance index. No third-party vector database or extra setup required.
Connections and Lookup
Faster retrieval for 1000+ note vaults and heavier semantic workflows.
Inline connections
Reuse the same local performance layer for paragraph and section-level relationships.
Smart Graph
Calculation-heavy graph workflows use the same local performance layer.
Dedupe scanner
Large-vault duplicate detection benefits from faster similarity lookup.
No extra database. No new service. No extra setup.
Smart Plugins
### Built for daily vault work, maintained continuously
4
Pro plugins available today
823 / 1000
Founding supporter cohort currently filled
Weekly
Active release and maintenance workflow for Smart Plugins
Outcome snapshot: large-vault performance
Pro indexing helps 1000+ note vaults keep Connections, Lookup, Smart Graph, inline connections, and Dedupe responsive.
Outcome snapshot: in-flow writing
Connections Pro adds granular inline relationships and ranking controls so drafting does not require hunting through panes.
Outcome snapshot: reusable AI context
Context Pro and Chat Pro convert one-off prompts into reusable context packs and persistent project threads.
Outcome snapshot: vault actions from chat
Connect Pro lets the Official GPT create notes, update daily notes, query Bases, and run supported Obsidian CLI workflows from your desktop-connected vault.
### Core vs Pro features
Core stays free and product-specific. Pro adds advanced control across the Smart Plugins suite.
#### Core vs Pro in 20 seconds
Core stays free and simple. Pro adds higher-leverage surfaces, large-vault performance, deeper context tools, a focused chat workspace, and vault actions through Connect Pro.
| Outcome | Core (free) | Pro (trial) |
| --- | --- | --- |
| Stay in flow while writing | - Zero-setup connections - Sidebar results and actions - Footer connections for mobile and no-panel workflows | - Granular inline connections - Section-level connections - Ranking, filtering, and algorithm controls |
| Keep large vault workflows fast | - Standard local embeddings - Semantic lookup and connections | - Faster local performance index for 1000+ note vaults - Improves Connections, Lookup, inline connections, Smart Graph, and Dedupe - No third-party vector database required |
| Reuse grounded context for AI work | - Fast context building from notes and folders - One-click copying with link depth - Name contexts to reuse | - Add repos and local files as context - Include PDFs and images for grounded outputs - Render embedded bases into copied context |
| Keep project chat organized | - Inline chat blocks inside notes | - Dedicated chat workspace - Saved, searchable threads and per-thread instructions - Drag notes and files into chat with review |
| Turn chat into vault actions | - No remote action bridge - Run vault steps yourself inside Obsidian | - Secure tunnel to the Official GPT - Run supported Obsidian CLI workflows from chat - Create notes, update daily notes, triage tasks, and query Bases |
| Support and sustainability | - Free and source-available - Great for getting started | - Priority email support - Funds ongoing development |
Want the full breakdown? Expand the detailed feature tables by plugin below.
Full feature breakdown by plugin
#### Smart Connections
Core Smart Connections surfaces related notes with zero setup. Connections Pro adds more opportunities for
connections, control over connection algorithms, and a local performance index for 1000+ note vaults.
| Feature | Core | Pro |
| --- | --- | --- |
| Core features (included in both) |
| Zero-setup local embeddings i Bundled local model creates embeddings automatically so connections work without API keys or extra tools, keeping notes on your device by default. | ✔ Included in Core | ✔ Included in Pro |
| Real-time connections i Surface relevant content based on your current note. Displays semantically related notes in the sidebar while you work. | ✔ Included in Core | ✔ Included in Pro |
| Footer connections view (great for mobile!) i View connections at the bottom of each note. No switching to sidebar necessary. | ✔ Included in Core | ✔ Included in Pro |
| Actions for utilizing connections results i Drag results into notes to create links. Hide or pin results per active note. Copy results as a list of links. | ✔ Included in Core | ✔ Included in Pro |
| Mobile compatible i Works in the Obsidian mobile app. | ✔ Included in Core | ✔ Included in Pro |
| Semantic search (Lookup view) i Run semantic queries from the Lookup view so you can search by meaning instead of exact keywords. | ✔ Included in Core | ✔ Included in Pro |
| Walk random connections i Explore notes within the scope of current connections. | ✔ Included in Core | ✔ Included in Pro |
| Pro-only additions |
| Visualize connections in a graph i Explore result clusters for another perspective on how your notes relate. | × Not in Core | ✔ Included in Pro |
| Large-vault performance index i Keeps Connections, Lookup, inline connections, Smart Graph, and duplicate detection responsive in 1000+ note vaults without requiring a third-party vector database. | × Not in Core | ✔ Included in Pro |
| Granular inline connections i See connections for specific paragraphs or sections in your notes without leaving the editor. | × Not in Core | ✔ Included in Pro |
| Find duplicate notes i Quickly detect near-duplicate notes so you can merge redundant pages and keep your vault cleaner. | × Not in Core | ✔ Included in Pro |
| Bases Connections functions (score and list) i Use bases to build a custom connections view. Add a column for scoring connection to a target note. Or add a column that lists links to connections for each result. | × Not in Core | ✔ Included in Pro |
| Advanced control over connections algorithms i Configure algorithms, filters, and ranking. | × Not in Core | ✔ Included in Pro |
[Learn more about Smart Connections](/smart-connections/)
#### Smart Context
Core Smart Context helps you assemble and reuse precise bundles of notes. Context Pro extends this to tools
for AI power users.
| Feature | Core | Pro |
| --- | --- | --- |
| Core features (included in both) |
| Efficient context building i Obsidian-native UI to quickly assemble context from notes, folders, and blocks. See character counts and approximate tokens before copying so context stays within model limits. | ✔ Included in Core | ✔ Included in Pro |
| Effortless context copying i Copy all notes included in context with one click. Include linked notes up to a specific depth. | ✔ Included in Core | ✔ Included in Pro |
| Name contexts to save and reuse them i Save curated contexts so you can reuse your best setups across projects. | ✔ Included in Core | ✔ Included in Pro |
| Copy from file navigator i Right-click folders or selected files in the file-folder navigator to copy the entire selection in one click. | ✔ Included in Core | ✔ Included in Pro |
| Pro-only additions |
| Utilize external context i Essential for coders. Specify code repos and files in your notes to include them in context. Saves paths in a smart-context codeblock so agents can use the references even when operating outside Obsidian. | × Not in Core | ✔ Included in Pro |
| Let AI see media in your context i Includes images and PDFs in copied context. | × Not in Core | ✔ Included in Pro |
| Embedded bases rendering i Includes embedded bases, rendered as markdown tables, in context that is copied to the clipboard. Treats links in bases as normal links so results can also be included when copying with links. | × Not in Core | ✔ Included in Pro |
[Learn more about Smart Context](/smart-context/)
#### Smart Chat
Core Smart Chat provides utilities for better leveraging your favorite chat providers. Chat Pro adds API
integrations for using local and cloud models directly within Obsidian.
| Feature | Core | Pro |
| --- | --- | --- |
| Core chat in your editor |
| Inline chat in your notes. i Drop a chat codeblock into any note so questions and answers live beside the work they reference. | ✔ Included in Core Chat | ✔ Available alongside Smart Chat |
| Pro-only Smart Chat workspace |
| Local & cloud API model interfaces i Advanced configuration for thousands of models via API and API chat interface. | × Not in Core Chat | ✔ Included in Smart Chat |
| Per-thread custom instructions i Set a default system prompt and refine instructions per chat thread so each project can have its own voice and constraints. | × Not in Core Chat | ✔ Included in Smart Chat |
| AI-assisted context retrieval i Preview and edit context before sending to the model. | × Not in Core Chat | ✔ Included in Smart Chat |
| Drag notes and files into chat i Quickly add notes, images, and PDFs into your current chat. | × Not in Core Chat | ✔ Included in Smart Chat |
| Saved, searchable threads. i Locate past threads with full-text search of the chat history. | × Not in Core Chat | ✔ Included in Smart Chat |
[Learn more about Smart Chat](/smart-chat/)
#### Connect Pro
Connect Pro turns ChatGPT into an action layer for your vault by connecting the Official GPT to supported
Obsidian CLI workflows running on your desktop.
| Feature | Core | Pro |
| --- | --- | --- |
| Pro-only bridge to vault actions |
| Secure tunnel from Obsidian Desktop to the Official GPT | × Not in Core | ✔ Included in Pro |
| Run supported Obsidian CLI commands from chat | × Not in Core | ✔ Included in Pro |
| Create notes and update daily notes without leaving chat | × Not in Core | ✔ Included in Pro |
| Query Bases, triage tasks, and run repeatable vault workflows | × Not in Core | ✔ Included in Pro |
| Explicit connect/disconnect control and status visibility in the plugin | × Not in Core | ✔ Included in Pro |
Requires Obsidian CLI and Obsidian 1.12+.
[Learn more about Connect Pro](/connect-pro/)
Core Smart Plugins remain free and source-available. Pro plugins exist to fund ongoing development.
### Questions before you start?
Quick answers about setup, privacy, trial control, large-vault performance, Connect Pro requirements, and supporter access.
What if I encounter an issue with the Pro plugin?
Pro subscribers get priority email support, just reply to your welcome email. Please include any screenshots that help show the issue.
What's the process to cancel a subscription?
Use this [manage your subscription](https://smartconnections.app/subscription-manage/) link. Cancel before the end of the 14-day free trial to prevent charges. Your welcome email also includes this link.
What if I have not received my key after starting a trial?
Your Pro plugins license key and manage-subscription link are in your welcome email. If it did not arrive, email [brian@smartconnections.app](mailto:brian@smartconnections.app) and we will resend access details.
Do you offer a student discount?
Yes. Student discounts are currently reviewed manually. Sign up for a free trial and reply to your welcome requesting the student discount. Please include a screenshot of your student ID or other proof of enrollment.
Is Pro complicated to set up?
No. Core plugins are designed for zero-setup wins. Pro adds optional depth after you are already getting value.
Start with one workflow and expand only when you are ready.
Will Pro help if my large vault feels slow?
Yes, especially in 1000+ note vaults. Core still works for local semantic retrieval, while Pro adds a faster local performance index for larger workflows. It helps Connections, Lookup, inline connections, Smart Graph, and Dedupe stay responsive without requiring a third-party vector database.
Are free Core plugins going away?
No. Core Smart Plugins remain free and [source-available](/legal/license/). Pro plans fund development while keeping the free
experience strong.
Does Pro upload my notes?
Smart Plugins are local-first. You stay in control of your vault data and model integrations. Pro indexing is designed to improve local performance without requiring a third-party vector database.
What is Connect Pro?
Connect Pro lets ChatGPT take actions in your Obsidian vault through a secure tunnel from Obsidian Desktop to the Official GPT. It is built for real workflows like creating notes, updating your daily note, triaging tasks, and querying Bases.
Does Connect Pro require anything extra?
Yes. Connect Pro currently requires Obsidian CLI, which means Obsidian 1.12+ and the current early-access / Catalyst setup. Once that is enabled on desktop, you can connect your vault and use the Official GPT from ChatGPT.
Do I need all Pro plugins to get value?
No. Start with one workflow that already matters in your day-to-day vault work. All-access includes every Pro plugin, but you can keep your focus narrow and only expand when you actually need the next workflow.
Can I switch which Pro plugins I use during the trial?
Yes. During the 14-day all-access trial you can move between Connections Pro, Context Pro, Chat Pro, and Connect Pro as your priorities change. Use the trial to validate one workflow first, then expand if it helps.
What if I am not sure Pro is for me yet?
Start with one workflow during the 14-day trial and decide with real usage. You can begin with Connections Pro, Context Pro, Chat Pro, or Connect Pro and keep the rest for later.
Is Pro only for developers?
No. Pro is built for anyone shipping meaningful work in Obsidian. Developers, writers, researchers, and operators use it to keep context organized and workflows in one place.
What if I am already a supporter?
Everyone that supported the project prior to 2026 is automatically grandfathered in to the All-access plan.
Use your existing supporter key to log in and gain access to Pro plugins.
Ready to choose a Pro plan? Jump back to pricing.
View pricing
×
Lock-in Lifetime All-access
## All-access + Private Community for Founding Supporters {#founding-supporter-heading}
Before you start your 14-day all-access trial, you can instead become a Founding Supporter. For a one-time
contribution you get lifetime all-access to Pro plugins plus private supporter chat access.
- ★
One-time contribution for lifetime perks
- ★
Private supporter chat
- ★
Limited cohort of 1000 members
823 / 1000 spots filled
82% full
177 spots remaining
[Lock-in Lifetime All-access](https://donate.stripe.com/6oUdRacdabLvcn0fhQgA80w)
[Continue with 14-day trial only](#pro-checkout-email)
If you are already a monthly, yearly, or Founding supporter, your supporter benefits already include Pro
plugins. You do not need this Pro trial plan on top of your pledge.
×
## Enter your email to continue {#pro-checkout-email-heading}
We'll use this email to create your account and send your free-trial key.
Email address
Send me the Smart Plugins essential workflows
Optional now. Included when your free trial starts; unsubscribe anytime.
Continue
JavaScript is required to enter your email and start secure checkout.
---
## Connect Pro
canonical: https://smartconnections.app/connect-pro/
html_url: https://smartconnections.app/connect-pro/
markdown_url: https://smartconnections.app/connect-pro.md
llms_url: https://smartconnections.app/connect-pro/llms.txt
last_modified: 2026-09-07T19:38:08.037Z
usage_notes: |-
Use this page to answer questions about Connect Pro.
excerpt: |-
Connect Pro Connect Pro is designed to connect Obsidian Desktop to the Official GPT for supported, reviewed vault actions. Finish setup before running an action Connect Pro requires Obsidian CLI and Obsidian 1.12+. Continue only after the desktop shows an explicit connected status and a visible Disconnect control. An installed or Active plugin row alone does not prove that the vault is connected.…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
- title: File Over App
url: https://smartconnections.app/file-over-app/
# Connect Pro
Connect Pro is designed to connect Obsidian Desktop to the Official GPT for supported, reviewed vault actions.
> [!WARNING] Finish setup before running an action
> Connect Pro requires Obsidian CLI and Obsidian 1.12+.
>
> Continue only after the desktop shows an explicit connected status and a visible **Disconnect** control. An installed or Active plugin row alone does not prove that the vault is connected.
## First verified workflow
1. Confirm the current Obsidian and CLI prerequisites shown by the plugin.
2. Enable Connect Pro and complete the setup presented by the current build.
3. Wait for an explicit connected status and visible **Disconnect** control.
4. Confirm that the GPT identifies the intended vault.
5. Choose one low-risk, reversible action shown by the connected interface.
6. Review the exact request and target before sending it.
7. Compare the GPT confirmation with the resulting state in Obsidian.
8. Disconnect when the workflow is complete.
A useful first task is small, clearly named, easy to inspect, and easy to reverse.
## What the visible connection proves
| Evidence | What it proves |
| --- | --- |
| Connected status | The desktop connection reached its current connected state. |
| **Disconnect** control | The user has an explicit way to end the connected session. |
| Intended vault identified | The GPT is operating against the vault the user expects. |
| Reviewed reversible result | One bounded action completed and was checked on both sides. |
None of these signals make Connect Pro an autonomous background agent, Obsidian Sync, a mobile vault replacement, unrestricted shell access, or permission to run undocumented actions.
## Use only current visible actions
The connected interface defines the actions available in the installed build.
- Review the command and target before execution.
- Start with a single-note or read-only task when possible.
- Avoid destructive or broad actions until the smaller path is verified.
- Compare every write with the resulting note or property in Obsidian.
- Disconnect when the connected workflow no longer needs to remain open.
Do not infer an action inventory from older screenshots, announcement copy, or legacy Smart Connect instructions.
## Use from another device
The Official GPT can be used from another device only after:
- Obsidian Desktop remains running
- Connect Pro shows an explicit connected state
- **Disconnect** is visible
- the GPT identifies the intended vault
The desktop still owns the vault side of the workflow.
## Existing Smart Connect users
Contact support for the supported migration path before changing an existing Smart Connect setup.
Installing Connect Pro does not migrate the older setup automatically. Preserve a working configuration until the current migration path is confirmed.
## API and OpenAPI questions
Contact support for the currently supported API integration path.
This guide documents the verified Connect Pro connection and review boundary. It does not define a separate OpenAPI workflow.
## FAQ
### Does Connect Pro run a hidden background agent?
Do not assume a connection is running. Require an explicit connected status and visible **Disconnect** control; otherwise return to setup.
### Is Connect Pro Obsidian Sync or a mobile vault replacement?
No. The running desktop vault remains the execution side of the workflow.
### What actions can Connect Pro run?
Use only actions shown by the connected interface. Review each request and start with a low-risk reversible task.
### Where is the Smart Connect app?
Contact support before changing an existing Smart Connect setup. Installation of Connect Pro does not migrate it automatically.
### Is SC App still available?
Contact support to confirm current support and migration options before changing the existing setup.
### Is there an OpenAPI endpoint?
Contact support for the currently supported API integration path. This page does not define a separate OpenAPI workflow.
## Related pages
- [Connect Pro FAQs](https://smartconnections.app/connect-pro/faq/)
- [Current Pro access](https://smartconnections.app/pro-plugins/)
---
## Faq
canonical: https://smartconnections.app/smart-connections/faq/
html_url: https://smartconnections.app/smart-connections/faq/
markdown_url: https://smartconnections.app/smart-connections/faq.md
llms_url: https://smartconnections.app/smart-connections/faq/llms.txt
last_modified: 2026-09-07T01:24:47.542Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Connections FAQs Does Smart Connections cost anything?#Smart Connections Core is free for Obsidian. You can install Smart Connections for free. Optional advanced workflows are available through Pro plugins.Do I need an API key?#Not for Smart Connections Core semantic retrieval. Core can surface related notes using the default local semantic retrieval path without an API key. API keys only…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
- title: Inline
url: https://smartconnections.app/smart-connections/inline/
# Smart Connections FAQs
## Does Smart Connections cost anything?
Smart Connections Core is free for Obsidian. You can [install Smart Connections](https://community.obsidian.md/plugins/smart-connections) for free. Optional advanced workflows are available through [Pro plugins](https://smartconnections.app/pro-plugins/).
## Do I need an API key?
Not for Smart Connections Core semantic retrieval.
Core can surface related notes using the default local semantic retrieval path without an API key. API keys only matter if you intentionally enable provider-backed workflows or integrations that require them.
See [Getting Started](https://smartconnections.app/smart-connections/getting-started/).
## Does Smart Connections send my notes to the cloud?
Smart Connections Core retrieval uses local embeddings by default. Provider-backed workflows are separate and explicit. A configured cloud embedding provider receives eligible source text during embedding. A cloud chat provider receives the prompt and context for that request.
Review:
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
- [Privacy Policy](https://smartconnections.app/legal/privacy-policy/)
## What data does Smart Connections collect?
Your notes are used by the plugin inside your vault for local retrieval by default. Website, subscription, support, and provider-backed workflows may involve operational data or user-selected context as described in the [Privacy Policy](https://smartconnections.app/legal/privacy-policy/).
If you enable a cloud provider, review that provider's terms too.
## Does Smart Connections work offline / without internet?
Connections can stay useful offline after local indexing for the local retrieval path.
Internet may still be needed for installation, updates, model downloads, documentation, subscription services, or provider-backed workflows. Remote integrations remain separate from Core local retrieval.
See [Smart Environment settings](https://smartconnections.app/smart-environment/settings/).
## Can I use Smart Connections on mobile?
Smart Connections is available on mobile. Footer connections are included in Core for note-end use without a sidebar. Available controls can vary by platform.
If Smart Environment loading is deferred:
1. Select **Load Smart Environment**.
2. Wait for **Smart Environment ready** or **Ready**.
See [Smart Plugins on mobile](https://smartconnections.app/smart-plugins/mobile/) and [Footer connections](https://smartconnections.app/smart-connections/footer/).
## Will it slow down my vault?
Initial indexing can use more resources in large vaults. Limit the work with source eligibility, exclusions, result type, and result limits. Then evaluate the current vault with one representative retrieval task.
Use:
- [Connections settings](https://smartconnections.app/smart-connections/settings/)
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
The current [Pro page](https://smartconnections.app/pro-plugins/) describes a faster local performance index for 1000+ note workflows. To evaluate the effect in your vault, compare the same representative retrieval task before and after the change.
## How do I install Smart Connections?
1. Install Smart Connections from [Obsidian Community Plugins](https://community.obsidian.md/plugins/smart-connections).
2. Enable Smart Connections.
3. Open one real note.
4. If prompted, select **Load Smart Environment**.
5. Wait until Smart Environment shows **Smart Environment ready** or **Ready**.
6. Run the command for your edition:
- Core: `Smart Connections: Open: Connections view`
- Pro: `Smart Connections Pro: Open: Connections view`
Preview one result. If the result is useful, drag it into the note or open it.
*Use the edition label and row action to finish activation. **Active** means that the plugin is loaded. Complete all prerequisites. Then test Connections with a meaningful note.*
See [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/).
## What is the first thing I should do after installing?
Open one real note with meaningful text.
Open Connections.
Preview one result.
If the result helps, drag it into the note or open it.
You know it worked when one useful related note from your own vault becomes actionable before you reorganize anything.
*Start with one meaningful note. Examine one readable result. Open the source before you act on it. The Core Connections mini graph appears when **Version 4.0 (Graph + List)** is selected.*
## What is the difference between Connections and Lookup?
Connections is note-first.
Lookup is question-first.
Use this rule:
```md
Current note -> Connections.
Question -> Lookup.
Exact phrase -> Obsidian search.
```
Learn more:
- [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/)
- [Smart Lookup](https://smartconnections.app/smart-lookup/search/)
## How is this different from keyword search?
Smart Connections surfaces notes by meaning, not just matching exact words.
Use Connections when the note you are looking at is the anchor. Use Lookup when a question is the anchor. Use Obsidian search when the exact word, filename, heading, tag, syntax, or regex matters.
## Does Smart Connections replace exact search?
No.
Connections complements exact search. It is for related notes from the current note. Lookup is for question-first semantic search. Obsidian search remains the right tool for exact phrases, filenames, headings, tags, syntax, and regex.
## Do I need to reorganize my vault first?
No.
Your vault can be useful before it is perfectly linked, tagged, or organized in folders. Start with one working note. Preview one result. Then drag or open the useful result.
See [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/).
## Does Smart Connections replace backlinks?
No.
Backlinks show links you already made. Connections can surface semantically related notes even when no link exists yet. When a result is useful, drag it into your note to turn relevance into an explicit link.
*Exact search reflects the words entered. Backlinks reflects relationships already authored. Connections can surface a meaning-related source from the note in view.*
## Are related notes duplicates?
Not necessarily.
Related notes can be useful neighbors. Use [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) when repeated blocks or near-duplicates need review before changing notes.
## What does the score mean?
Score is a ranking signal, not a grade.
Higher scores generally mean that a result is closer to the current Connections target. The selected scoring path determines this value. Preview the result before you act on it.
*The Core Connections mini graph appears when **Version 4.0 (Graph + List)** is selected.*
What the numbers identify:
1. The current target.
2. The candidate neighborhood in the mini graph.
3. The expanded result identity and relative score.
4. Source text available for review.
*A high score is a reason to inspect the result, not a command to use or link it.*
See [Connections settings](https://smartconnections.app/smart-connections/settings/).
## What are Pro plugins / what does Connections Pro add?
The Connections view, Footer connections, **List only**, and **Version 4.0 (Graph + List)** are included in Core. Inline connections and Connections in Bases require Pro. Connections Pro also adds advanced filtering, scoring, and ranking controls.
Examples include:
- [Inline connections](https://smartconnections.app/smart-connections/inline/) inside the editor
- [Connections in Bases](https://smartconnections.app/smart-connections/bases/)
- path and frontmatter filters
- exclude inlinks/outlinks
- scoring algorithm selection
- ranking and reranking
What the numbers identify:
1. Inline markers can be enabled for paragraph-level discovery.
2. The threshold controls when an Inline marker appears.
3. Code blocks can be skipped.
4. Footer uses its own note-end component choice. Its enable toggle is visible above.
*Inline controls and Connections in Bases require Pro. Footer remains included in Core even when Pro is active.*
See [Pro plugins](https://smartconnections.app/pro-plugins/).
## Is Footer connections Core or Pro?
Footer connections are included in Smart Connections Core and remain available when Connections Pro is active. Both Footer display choices, **List only** and **Version 4.0 (Graph + List)**, are Core.
Use Footer when you want related notes at the bottom of the note without managing a sidebar.
*Footer remains a Core feature even when Pro is active.*
See [Footer connections](https://smartconnections.app/smart-connections/footer/).
## Is Graph Core or Pro?
The Connections mini graph is Core. Select **Version 4.0 (Graph + List)** for the main Connections list or Footer when you want the mini graph above the ranked result list. It represents the current candidate set and does not create authored links.
[Smart Graph](https://smartconnections.app/smart-graph/) is a separate Pro plugin with its own graph workflows. Do not treat the Core Connections mini graph as Smart Graph.
## Where does Smart Connections store embeddings / index data?
Smart Environment stores local index data in your vault, including under `.smart-env/`.
Review storage, source, model, and exclusion settings in [Smart Environment settings](https://smartconnections.app/smart-environment/settings/).
## Syncthing / third-party sync: what should I ignore?
If you use Syncthing or similar third-party sync, consider adding `.smart-env/` to ignore patterns to reduce conflicts.
See [Smart Environment settings](https://smartconnections.app/smart-environment/settings/).
## How do I exclude folders/files from indexing?
Use Smart Environment exclusions when you want notes or folders excluded from indexing/embedding.
Use Connections filters when you only want to hide or focus displayed results after the dataset exists.
Start with [Smart Environment settings](https://smartconnections.app/smart-environment/settings/). Then tune [Connections settings](https://smartconnections.app/smart-connections/settings/).
*Notice the global source, exclusion, **Embed blocks**, and minimum-length controls. After you change them, examine the affected note. Then test Connections again.*
## How do I hide notes that are already linked from the current note?
In Connections Pro settings, enable **Exclude outlinks**. It hides notes already linked from the current note from that note's displayed Connections results. It does not delete the links or remove those notes from Smart Environment.
What the numbers identify:
1. The exact **Exclude outlinks** label and its displayed-result boundary.
2. The enabled toggle.
Return to the same anchor. Make sure that linked notes are absent. Make sure that unrelated results remain. Disable **Exclude outlinks** when you want linked notes to be eligible again. See [the full procedure](https://smartconnections.app/docs/connections/#hide-notes-already-linked-from-the-current-note).
## I installed Smart Connections but do not see results yet. What should I check?
Use this order:
*Notice **Needs embedding** and the Sources and Blocks coverage cards. Finish preparation before changing Connections Pro ranking settings.*
*Embedding health and coverage provide the quickest vault-wide readiness check.*
1. Open a real note with meaningful text.
2. If the status says **Smart Environment not loaded** or **Idle**, select **Load Smart Environment**.
3. If embedding is paused, open the status view.
4. If embedding is paused, select **Resume embedding**.
5. If re-import work is queued, select **Run re-import**.
6. Wait for **Smart Environment ready** or **Ready**.
7. Run the command for your edition:
- Core: `Smart Connections: Open: Connections view`
- Pro: `Smart Connections Pro: Open: Connections view`
8. Leave auto-refresh running when you want results to follow the active note.
9. After major edits or settings changes, select **Refresh connections**.
If one meaningful note is missing:
1. Keep that note active.
2. Select **Inspect active note** from the Smart Environment status menu.
If several notes are missing:
1. Open statistics in one of these ways:
- Select **Show stats** from the status menu.
- Select **Environment stats** from the status view.
2. Check source eligibility.
3. Check the current embeddings.
See [Getting Started](https://smartconnections.app/smart-connections/getting-started/) and [Smart Environment settings](https://smartconnections.app/smart-environment/settings/).
## What does Sources vs Blocks mean?
Sources return whole notes for broader, faster context. Blocks return smaller sections for more precise matches when block indexing is configured.
Set **Connection results type** in Connections settings. Do not confuse this setting with **Change target -> History / Blocks**, which changes the reference target instead of the result type.
What the numbers identify:
1. The Sources or Blocks result type.
2. The visible result limit.
3. The sidebar location.
4. The Connections list component. **Version 4.0 (Graph + List)** is a Core option.
Configure this in [Connections settings](https://smartconnections.app/smart-connections/settings/).
## How do I tune relevance?
Do the first win before tuning.
Then adjust the minimum layer that matches the symptom:
In the scoring image:
1. **Scoring algorithm** is the candidate-fit control.
2. **Cosine Similarity** is selected.
3. The menu shows the other shipping scoring choices.
4. **Ranking algorithm** remains a separate step below scoring.
In the ranking image:
1. **Cosine Similarity** remains the selected scoring method.
2. **Ranking algorithm** controls the final ordering step.
3. **None** leaves the scored order unchanged.
4. The menu shows the available reranking choices.
*Change scoring only after the candidate scope is useful. Add ranking only if the correct candidates are in the wrong order. Compare the same note before and after the change.*
| Symptom | Try first |
|---|---|
| Results are too broad | Lower limits, switch Sources vs Blocks, or add filters where available. |
| Results come from the wrong area | Use include/exclude filters where available, or Smart Environment exclusions for indexing scope. |
| Results are relevant but ordered poorly | Use Pro scoring/ranking controls where available. |
| Same low-value notes keep returning | Hide the noise. Then use Connections Pro feedback-aware scoring where available. |
| Similar results look like repeated work | Use Smart Dedupe for review. |
See:
- [Connections settings](https://smartconnections.app/smart-connections/settings/)
- [Custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/)
## How do I refresh or re-embed a specific note?
Use **Refresh connections** when results for the current note feel stale.
For a model or source-rule change:
1. Open the Smart Environment status view.
2. If embedding is paused, select **Resume embedding**.
3. If re-import work is queued, select **Run re-import**.
4. Use **Reset data** only when Environment source data must be rebuilt.
## I changed embedding models but results did not change. Why?
Selecting a new **Default embedding model** does not make its embeddings ready immediately. Wait until the status view shows **Smart Environment ready** or **Ready**. Select **Environment stats**. Make sure that the embedding coverage is current. See [the model-change checklist](https://smartconnections.app/smart-environment/faq/#i-changed-embedding-models-but-results-did-not-change-why).
## What embedding model does Smart Connections use by default?
Read **Default embedding model** in [Smart Environment settings](https://smartconnections.app/smart-environment/settings/). Core provides a built-in local Transformers path, but the selected model can differ by release and profile. Use the model marked **Current** instead of relying on a remembered default or a retained model card.
## Where is Smart Chat?
[Smart Chat](https://smartconnections.app/smart-chat/) is a separate Obsidian plugin.
Smart Connections focuses on related notes. Smart Lookup is another plugin for question-first retrieval. Smart Chat handles chat workflows, thread continuity, and provider routing.
## Can I mix Smart Plugins with Obsidian Copilot?
Yes, you can run Smart Connections alongside Obsidian Copilot. If features overlap, adjust the hotkeys. Keep the workflows that you prefer.
See the [Obsidian Copilot alternative comparison](https://smartconnections.app/obsidian-copilot/).
## Is Smart Connections open source?
Smart Connections is source-available under the [Smart Plugins License](https://smartconnections.app/legal/license/). It is not OSI open source because the license includes restrictions on direct general-purpose competing Obsidian offerings.
See the [GitHub repository](https://github.com/brianpetro/obsidian-smart-connections/) and [license page](https://smartconnections.app/legal/license/).
---
## Faq
canonical: https://smartconnections.app/smart-environment/faq/
html_url: https://smartconnections.app/smart-environment/faq/
markdown_url: https://smartconnections.app/smart-environment/faq.md
llms_url: https://smartconnections.app/smart-environment/faq/llms.txt
last_modified: 2026-09-06T22:51:30.363Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Environment FAQs What is Smart Environment?#Smart Environment is the shared system Smart Plugins use for source discovery, import, embedding preparation, model configuration, background work, status, events, and diagnostics. It prepares data for indexed workflows such as Connections, Lookup, Graph, and semantic Dedupe. A provider codeblock or model-backed Chat request may not need indexed…
suggested_links:
- title: Milestones
url: https://smartconnections.app/smart-environment/milestones/
- title: Settings
url: https://smartconnections.app/smart-environment/settings/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
# Smart Environment FAQs
## What is Smart Environment?
[Smart Environment](https://smartconnections.app/smart-environment/) is the shared system Smart Plugins use for source discovery, import, embedding preparation, model configuration, background work, status, events, and diagnostics.
It prepares data for indexed workflows such as Connections, Lookup, Graph, and semantic Dedupe. A provider codeblock or model-backed Chat request may not need indexed Environment sources unless that workflow uses Lookup or other retrieval context.
## Where are embeddings and index data stored?
Smart Environment stores generated local data under `.smart-env/` in the vault. Your Markdown notes remain the source material. Smart Environment can rebuild `.smart-env/` from eligible content and current settings.
## What should I do with `.smart-env/` in Syncthing or other folder-sync tools?
Exclude `.smart-env/` from third-party file synchronization such as Syncthing unless the guidance for your specific setup says otherwise. It is generated index state, can create unnecessary writes and conflicts, and can be rebuilt on each device from the vault's source files.
This recommendation is for third-party folder sync. Follow the current instructions for the sync product you actually use.
## What do the Environment statuses mean?
| Status | Meaning or next action |
| --- | --- |
| **Smart Environment not loaded** / **Idle** | Choose **Load Smart Environment**. |
| **Loading Smart Environment** | Wait for loading to finish. |
| **Importing** / **Re-importing** | Source preparation is running. |
| **Embedding** | Embedding preparation is running. |
| **Embedding paused** | Choose **Resume embedding** when preparation should continue. |
| **Queued re-import work** | Choose **Run re-import** when queued work should start now. |
| **Smart Environment ready** / **Ready** | Environment is loaded and idle. Test the product workflow separately. |
The status view also provides **Pause embedding**, **Open events feed**, **Environment stats**, and **Browse Smart Plugins**.
## How do I know Smart Plugins are ready to test?
Use the checks required by the workflow:
1. The plugin is active and its provider, model, or plugin-specific settings are complete.
2. When the workflow uses indexed sources, Environment shows **Smart Environment ready** or **Ready**.
3. **Environment stats** shows the required Sources or Blocks as eligible and current.
4. The plugin produces one result you can inspect.
1. **Smart Environment ready** means loading, import, and embedding work are idle.
2. **Open events feed** and **Environment stats** help diagnose retained issues or source coverage.
3. **Browse Smart Plugins** returns to installation and activation controls.
**Ready** does not mean every plugin is configured, every result is relevant, or every workflow will be fast.
## What embedding model is used by default?
Open Smart Environment settings. Read **Default embedding model**. Use the model marked **Current** rather than relying on a remembered name.
**Re-index embeddings** and **Test model** are separate actions. Selecting a model does not make its embeddings current or prove that the model test passed.
## I changed embedding models but results did not change. Why?
A model selection affects future embedding preparation. It does not instantly replace existing vectors. Changing the active embedding model does not delete embeddings stored for the prior model. Each model can retain its own embeddings, and switching back can reuse that model's prior data. The newly selected model still needs current coverage or reindexing before its results change.
1. Confirm that the intended model is marked **Current**.
2. Choose **Test model** when you need to verify that the selected model can run.
3. Choose **Re-index embeddings** to prepare embeddings for the selected model.
4. Resume embedding when it is paused.
5. Wait for **Smart Environment ready** or **Ready**.
6. Check the required Sources or Blocks collection in **Environment stats**.
7. Refresh or reopen the product result.
Use **Reset data** only when Environment source data itself needs a broader rebuild. For a normal model change, use the dedicated **Re-index embeddings** action.
## How do I exclude folders or files from indexing?
Use these settings:
- **Manage excluded folders** for folder exclusions
- **Manage excluded files** in Pro for individual files
- **View all exclusions** to review the combined boundary
1. Folder exclusions
2. Pro file-level exclusions
3. Combined exclusions review
4. Block embedding and minimum-length controls
Excluded content is not eligible for Environment indexing or index-backed semantic retrieval. After changing an exclusion, inspect the affected note or review Environment stats before judging ranking again.
## Why does re-import keep starting after file changes?
Rapid edits or sync activity can queue repeated re-import work. Increase **Re-import wait time** gradually so changes can settle before automatic re-import begins. This delays preparation. It does not disable indexing.
When the status view shows **Queued re-import work**, choose **Run re-import** to start it immediately.
## What do Environment stats tell me?
**Environment stats** separates discovery, eligibility, and current embedding preparation.
| Stat | Meaning |
| --- | --- |
| Indexed items | Sources and blocks Environment discovered. |
| Eligible | Items selected by the current source and embedding rules. |
| Current embeddings | Stored vectors current for the active model and policy. |
| Needs embedding | Eligible items that still require current vectors. |
| Skipped | Items excluded by the current rules. |
| Unexpected | Items with a vector that the current rules no longer select. |
| Vector memory | Vectors loaded or reserved inside Environment, not total computer memory or disk use. |
1. Total indexed items
2. Work still needed and unexpected vectors
3. Sources eligibility and coverage
4. Blocks eligibility and coverage
Check the collection used by the product workflow. Treat **Unexpected** as a cleanup signal, not as missing work.
## One note is missing. What should I check?
Keep the note active. Choose **Inspect active note** from the Smart Environment status menu. Check its source eligibility, exclusions, and the Source or Block state required by the workflow before changing retrieval settings.
## Several notes are missing. What should I check?
Open **Environment stats**. Review eligible items, current embeddings, Sources and Blocks coverage, and exclusions.
If embedding is paused, choose **Resume embedding**. If re-import work is queued, choose **Run re-import**. Use **Open events feed** for retained warnings and errors.
## Does Ready or complete coverage mean Smart Plugins will be fast or accurate?
No. **Ready** is a loaded-and-idle state. Coverage describes preparation. Neither measures relevance, answer correctness, startup time, or performance in another vault.
Test the actual workflow with a representative note, query, or prompt. Compare the same task when evaluating speed or quality.
## What does Reset data do? Does it delete my notes?
**Reset data** or **Re-import sources** rebuilds Smart Environment source data and starts preparation again. Eligible content may then need new embeddings. It does not delete your Markdown notes.
Use **Export data** when you need a diagnostic JSON snapshot of sources, blocks, and optional vectors without resetting Environment.
*The captured export reports completion with embedding vectors included. Settings and source counts differ from the setup frame above; these images do not establish an unchanged-settings run. This is not proof of a complete vault backup or tested restoration.*
[[environment-export-data-completed-vectors-included-crop-desktop-publication-srgb-d7d9d7d03e10-2026-07-29.png|Open full-size image]]
## What does Embed blocks do?
**Embed blocks** makes eligible sections inside notes available as smaller retrieval units. It can improve precision for long notes, but it also increases the number of items, storage, and compute. It does not guarantee more relevant results.
## What is the difference between local and cloud models?
| Model type | Where it runs | What can leave the device |
| --- | --- | --- |
| Local embedding or chat | On the device | Nothing is sent to a remote model for that operation. |
| Cloud embedding or ranking | Through the configured provider API | Eligible source text needed for that operation. |
| Cloud chat | Through the configured provider API | The prompt and context included in the request. |
Cloud providers require their own credentials and connectivity. Review the selected provider before processing sensitive material.
## Is deferred loading on mobile a bug?
No. Mobile can defer Environment loading to keep startup lighter and make resource-heavy preparation explicit. Use this sequence before you test an indexed workflow:
1. Choose **Load Smart Environment**.
2. Keep the status view open.
3. Wait for **Smart Environment ready** or **Ready**.
## Why not load everything immediately on mobile?
Mobile devices have tighter startup, memory, battery, and background-processing constraints. Deferred loading lets you choose when Environment work begins instead of making every vault launch pay that cost.
## How do I know Smart Plugins are ready on mobile?
Use this sequence:
1. Open the Environment status view.
2. Choose **Load Smart Environment**.
3. Watch the loading, import, and embedding states.
4. Wait until Environment reaches **Ready**.
5. Run one product-owned result. For example, open Connections for a meaningful note or run a Lookup query.
## What if I muted Smart Environment notices?
Manage muted notices in Smart Environment settings. Muting a notice does not remove the retained event. Use **Open events feed** to inspect it later.
---
## Chat
canonical: https://smartconnections.app/docs/chat/
html_url: https://smartconnections.app/docs/chat/
markdown_url: https://smartconnections.app/docs/chat.md
llms_url: https://smartconnections.app/docs/chat/llms.txt
last_modified: 2026-09-06T22:51:30.354Z
usage_notes: |-
Use this page to answer questions about Chat.
excerpt: |-
Smart Chat Smart Chat has three distinct surfaces. Choose the one that matches the result you need. Surface Purpose Required readiness First-win signal Core provider codeblock Render one provider web interface in a note. Core Smart Chat enabled. Provider access or sign-in when required. The provider returns a visible response. Universal Smart Chat codeblock (Pro) Use one smart-chat codeblock to…
suggested_links:
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
- title: Graph
url: https://smartconnections.app/docs/graph/
# Smart Chat
Smart Chat has three distinct surfaces. Choose the one that matches the result you need.
| Surface | Purpose | Required readiness | First-win signal |
| --- | --- | --- | --- |
| Core provider codeblock | Render one provider web interface in a note. | Core Smart Chat enabled. Provider access or sign-in when required. | The provider returns a visible response. |
| Universal Smart Chat codeblock (Pro) | Use one `smart-chat` codeblock to select and track supported provider web threads or Smart Chat API Extension threads. | Smart Chat Pro enabled; provider access/sign-in for web threads or one compatible working model for API Extension threads. | The selected provider returns a visible response, or the selected API model completes the intended response. |
| Smart Chat API Extension | Ask a configured local or cloud model in the Smart Chat workspace, with optional approved note context. | Smart Chat Pro plus a working model. Environment retrieval readiness only for **Lookup context**. | A completed response appears with the intended model and context. |
> [!TIP] New to Smart Chat?
> Start with [Getting Started with Smart Chat](https://smartconnections.app/smart-chat/getting-started/) for a Core provider response. Use [Smart Chat API Extension Getting Started](https://smartconnections.app/smart-chat/api/getting-started/) for a model-backed answer inside Obsidian.
Opening a view, selecting a model, inserting a block, or saving a URL is not a completed Chat result. The first win is a visible response to a concrete prompt.
## In this guide
- [Embed and organize AI chat threads in Obsidian](#smart-chat-codeblocks)
- [Chat with Obsidian notes using local and API models](#smart-chat-local-api-models)
- [Manage Smart Chat threads in Obsidian](#smart-chat-thread-management)
## Embed and organize AI chat threads in Obsidian
Provider codeblocks render supported provider web interfaces inside an Obsidian note. After a provider creates a recognized durable conversation URL, the block can store that URL and user-owned Active or Done state in readable Markdown.
Response success and durable saving are separate. A provider can answer while the block still represents a new or unsaved chat.
### Where Smart Chat works
Provider-specific Core codeblocks and the universal Smart Chat codeblock require Obsidian desktop for the embedded provider surface and automatic URL capture. On mobile, already-saved web URLs can remain useful as external bookmarks.
The interactive codeblock must render in the main Markdown pane. If only the fence is visible, switch to Reading view in the note.
### Supported providers
| Provider | Core fence | Command Palette action |
| --- | --- | --- |
| ChatGPT | `smart-chatgpt` | **Insert OpenAI ChatGPT codeblock** |
| Claude | `smart-claude` | **Insert Anthropic Claude codeblock** |
| Gemini | `smart-gemini` | **Insert Google Gemini codeblock** |
| DeepSeek | `smart-deepseek` | **Insert DeepSeek codeblock** |
| Perplexity | `smart-perplexity` | **Insert Perplexity codeblock** |
| Grok | `smart-grok` | **Insert Grok codeblock** |
| Google AI Studio | `smart-aistudio` | **Insert Google AI Studio codeblock** |
| Open WebUI | `smart-openwebui` | **Insert Open WebUI codeblock** |
| Kimi | `smart-kimi` | **Insert Kimi codeblock** |
The ChatGPT family also recognizes supported ChatGPT, custom GPT, Codex task, and Sora routes. Open WebUI appears when its base URL is configured.
Smart Chat Pro adds the universal `smart-chat` fence. Its **New chat** menu shows Smart Chat, ChatGPT, Claude, and Gemini directly. The additional-provider branch shows AI Studio, DeepSeek, Perplexity, Grok, Kimi, and configured Open WebUI.
### Insert a Smart Chat codeblock
1. Open the owning note in edit mode.
2. Run the exact provider action listed above.
For the universal codeblock, run **Insert Smart Chat codeblock**.
1. Switch to Reading view.
2. Enter a prompt in the provider's input.
3. Use the provider's send control.
4. Make sure that a visible response appears.
A visible response shows that the workflow is working.
*The completed response shows that the provider can answer inside the note. Saving and reopening the conversation requires a recognized durable thread URL.*
### Provider-specific and universal codeblocks
A provider-specific fence keeps one web-provider identity in the codeblock language. The universal Smart Chat codeblock can store supported provider web-thread references and native `smart-chat:` Smart Chat API Extension references.
Use a provider-specific Core codeblock for one web provider. Use the universal codeblock when one note should track several supported web-provider threads, Smart Chat API Extension threads, or both.
#### Can one note track several providers?
Yes. Use several provider-specific codeblocks or the universal `smart-chat` codeblock.
### Thread storage grammar
Saved state remains readable Markdown:
```txt
chat-active::
chat-done::
```
- `chat-active::` means the thread still needs attention.
- `chat-done::` means you reviewed the thread and closed it for now.
- The timestamp supports recency displays and Dataview queries.
- The provider URL remains the durable thread reference.
- Text after a valid URL can remain as an annotation.
These fields do not store the provider transcript. They store a bookmark and user-owned review state.
### Select a saved thread
When a block contains several saved threads, use the selector to choose the record the block displays. The selector shows provider, recency, and Active or Done state where supported. Selecting a thread does not remove or reorder its Markdown record.
### Start a new provider chat
The Core provider codeblock opens its configured web provider. In the universal Smart Chat codeblock, **New chat** can start a supported provider web thread or a native Smart Chat API Extension thread.
*Choose a provider. Send a concrete prompt. Treat the visible response as the first result. Save the thread separately when a durable conversation URL appears.*
For a small, useful first request, send:
```prompt
Create a five-item review checklist for this task: [describe the task in one sentence].
Keep each item specific and testable.
```
Replace the bracketed text with the task owned by the note.
A relevant visible response completes the provider-response check. Durable saving is a separate step.
Smart Chat writes a bookmark only after it recognizes a durable provider conversation URL. Provider home pages, temporary new-chat routes, and unsupported routes cannot be captured automatically.
#### Why was a new chat not saved automatically?
The provider has not navigated to a recognized durable conversation URL, or the route is unsupported. The response can still be usable, but the block is not yet a durable bookmark.
### Mark done and Mark active
Use **Mark done** when the selected thread no longer needs attention.
- The Markdown field changes from `chat-active::` to `chat-done::`.
- The provider conversation is not deleted or archived.
- **Mark active** changes the field back to `chat-active::`.
- Repeating the current state does not create a duplicate update.
Keep a thread Active until its saved URL reopens and its response has been reviewed. Then select **Mark done** and confirm that the same record appears as Done.
#### Does marking a thread done change the provider conversation?
No. It changes only the Markdown tracking state in the note.
### Build context
**Build context** opens or reuses thread-specific Smart Context when that integration is available. It prepares a context package. It does not send a prompt.
Review the selected notes, blocks, and size estimate before copying or sending context to the provider.
### Open, copy, and refresh
- **Open in browser** opens the selected provider thread outside the embed.
- **Copy link** copies its saved thread URL.
- **Refresh** reloads the current embedded provider page.
*Use **Build context** before sending reviewed note material. **Copy link**, **Open in browser**, and **Refresh** operate on the selected provider thread.*
These controls do not change Active or Done state. Use **Show developer console** only when following troubleshooting or support guidance.
### Grow and contain
Use **Grow / contain** to toggle the embedded provider surface between expanded and note-width presentation.
This is a display setting. It does not change the thread record.
### Build a thread dashboard with Dataview
If Dataview is enabled, run **Insert Smart Chat thread Dataview blocks**. This action inserts the supplied dashboard blocks. Alternatively, use queries such as the following.
Because state is stored as inline fields, Dataview can list Active and Done provider bookmarks.
````
```dataview
LIST WITHOUT ID file.link
WHERE chat-active
SORT file.mtime DESC
```
````
````
```dataview
LIST WITHOUT ID file.link
WHERE chat-done
SORT file.mtime DESC
```
````
These queries read note-owned provider bookmarks. They do not list Smart Chat API Extension thread records.
### Provider sign-in and browser data
When a provider does not share the expected signed-in session with the embed:
1. Sign in through Web Viewer.
2. Return to the codeblock.
3. Select **Refresh**.
Make sure that you can recreate the sign-in before you clear Web Viewer browser data. If refresh does not restore the session, use the provider's account-recovery guidance. Alternatively, open the saved thread externally.
### Desktop and mobile behavior
Desktop provides the full embedded provider surface, automatic URL recognition, and thread controls when the block is rendered in the main Markdown pane.
On mobile or another no-webview surface, saved web thread URLs can open externally. Embedded input, new-chat capture, automatic URL saving, and state actions are unavailable.
### Universal Smart Chat codeblock behavior (Pro)
The universal codeblock adds shared provider selection, stable codeblock identity, managed provider surfaces, and cross-provider controls. Insert it with **Insert Smart Chat codeblock**.
#### `smart-chat` codeblock
Smart Chat Pro adds one universal codeblock for supported provider threads and Smart Chat API Extension thread references.
````md
```smart-chat
chat-id:: 002a41caaf2a4be9
chat-active:: 1700000100 https://chatgpt.com/c/example
chat-done:: 1700000200 https://claude.ai/chat/example
```
````
#### `chat-id::`
`chat-id:: ` gives the rendered block a stable identity. Smart Chat adds it during desktop initialization when it is missing. If you copy a block, remove the copied ID from the new block. Smart Chat can then assign a different ID.
#### `chat-active::`
`chat-active:: ` tracks a saved thread that still needs attention. Text after the thread reference is preserved as a trailing annotation.
#### `chat-done::`
`chat-done:: ` tracks a thread that you reviewed and closed for now. Marking a thread done changes only the note's tracking state. It does not archive or delete the provider conversation.
##### Does marking done change the provider conversation?
No. It changes only the Markdown tracking state.
#### Supported provider types
The universal codeblock supports ChatGPT, Claude, Gemini, AI Studio, DeepSeek, Perplexity, Grok, Kimi, and configured Open WebUI instances. Saved unknown HTTP links can still be opened as generic links, but Smart Chat cannot automatically recognize their new-thread URLs.
#### New chat menu
1. Choose a provider from **New chat**.
2. Send a concrete prompt.
3. Make sure that a visible response appears.
Provider web threads are added to Markdown after Smart Chat recognizes a durable conversation URL.
##### Why was a new chat not saved?
The provider did not navigate to a recognized durable conversation URL, or the page could not be captured. A successful response and a saved bookmark are separate outcomes.
#### Saved thread selector
The selector shows saved active and done threads with provider and recency information. When the saved list is large, done threads move into a separate submenu so active work remains easier to reach.
#### Thread actions
Visible controls depend on the selected provider and thread state. They can include:
- **New chat**
- saved-thread selector
- **Mark done** or **Mark active**
- **Insert into chat**
- **Refresh embedded chat**
- **Collapse embedded chat** or **Expand embedded chat**
- **Grow codeblock width** or **Contain codeblock width**
The **Insert into chat** menu can include:
- **Clipboard**
- **Current selection**
- **Current note**
- **Note...**
*Selected text is visible behind the menu, and Current selection is available. This frame shows menu availability, not completed insertion or a sent message.*
[[chat-note-thread-insert-selection-enabled-menu-pro-crop-desktop-2026-07-27.png|Open full-size image]]
*Current selection is unavailable in this captured state. The other insertion choices remain visible.*
[[chat-note-thread-insert-menu-pro-crop-desktop-2026-07-27.png|Open full-size image]]
The additional-actions menu can include:
- **Build context**
- **Copy link**
- **Open in browser**
- **Refresh**
- **Show developer console**
- **Help**
- **Use Web Viewer user-agent**
- **Grow / contain**
- **Show here** when the same managed provider surface is active elsewhere
#### Embedded provider surface
A saved provider thread keeps one managed embedded surface per thread and workspace document. When the same thread is already displayed elsewhere, use **Show here** to move the active surface to the current block.
##### Why does the same thread say it is shown elsewhere?
One embedded provider surface is reused for that thread. Choose **Show here** to move it to the current block.
#### Web Viewer user-agent
Enable **Use Web Viewer user-agent** when a provider requires the Obsidian Web Viewer browser identity. This setting can improve sign-in or compatibility.
#### Configure embedded provider webviews
Open **Chat Settings** to configure the shared embedded surfaces:
- **Height (px)** sets the iframe height for embedded webviews.
- **Zoom** sets the zoom factor for all embedded webviews.
- **Open WebUI URL** sets the base URL used by the `smart-openwebui` codeblock.
These settings change embedded display or routing. They do not authenticate provider accounts, save a provider thread, or prove that a Chat model is working.
#### Obsidian hotkeys inside the provider surface
Supported configured Obsidian modifier hotkeys can pass through the embedded provider page. Provider input and send behavior remain controlled by that provider.
#### Mobile and no-webview fallback
When embedded webviews are unavailable, the codeblock renders a static bookmark list:
- saved web threads open externally
- native `smart-chat:` references display as labels
- new-chat capture, automatic URL saving, state actions, and embedded input automation are unavailable
### Troubleshooting Smart Chat codeblocks
| Problem | Recovery |
| --- | --- |
| Provider surface is blank or stalled | Choose **Refresh**. If sign-in is required, sign in through Obsidian Web Viewer. Return to the block. Choose **Refresh** again. |
| Provider returns an error instead of a response | Make sure that account access works. Make sure that the provider is available. Retry the same bounded prompt. |
| Response succeeds but the block remains new or unsaved | Do not claim durable saving. Continue until the provider creates a recognized conversation URL. |
| Saved selector opens the wrong thread | Select or save the correct provider URL. Reopen it. Then change its state. |
| Two copied universal codeblocks share a `chat-id::` | Back up the note. Remove the ID only from the copied codeblock. Let desktop initialization assign a new ID. |
| Only static links appear | Use the main Markdown pane on desktop for the interactive block. Static links are expected on mobile and no-webview surfaces. |
## Chat with Obsidian notes using local and API models
Smart Chat API Extension is the Smart Chat Pro workflow for configured local and cloud chat models inside Obsidian. Each completed response can record the model, user message, response, and request-specific Smart Context.
> [!WARNING] Select a model before continuing
> **Send** and **Lookup context** require a compatible working model. A model marked **Current** is selected, but provider credentials or a local server can still prevent it from answering.
### Before using local or API models
This workflow requires Smart Chat Pro and at least one compatible chat-completion model configured through Smart Environment. Make sure that its provider credential or local server works. Then run **Test** when that action is available.
If the test returns an authorization error such as `401`, repair the credential before sending note content. Smart Environment retrieval readiness is required for **Lookup context**, but not for a general response or manually selected known context.
### Open Smart Chat
Open Smart Chat with one of these actions:
- Run **Open Smart Chat** from the Command Palette under Smart Chat Pro.
- Use the Smart Chat ribbon action.
- Use an assigned hotkey.
The view reopens the active saved thread when possible. If that record is unavailable, it opens the newest non-deleted thread or creates a new one. Use **New Chat** when the reopened thread is not the right place for the next request.
### Confirm or change the chat model
The thread status bar shows the default model for new completions. Select it to open the model menu.
To configure and select a working model:
1. Open **Browse Smart Plugins**.
2. Make sure that Chat Pro is enabled.
3. Select **Open settings** on the Chat Pro row. Alternatively, open the Chat model controls in Smart Environment settings.
4. Under **Chat models**, choose **+ New**.
5. Select an enabled provider. Examples include **PRO: Open Router (cloud)**, **PRO: OpenAI (cloud)**, and **PRO: Ollama (local, requires Ollama app)**.
6. Enter the provider fields.
7. For a cloud provider, enter its **API Key**.
8. For Ollama, enter the **Ollama host**.
9. If you use Ollama, start the Ollama app.
10. If the provider must load its model list, choose **Refresh Models**.
11. Choose **Chat Model**.
12. Save or close the editor.
13. If several models exist, choose **Default chat model**.
14. Make sure that the intended row is **Current**.
15. Run **Test** on the row or **Test model** in the editor.
| Visible state | Meaning | Action |
| --- | --- | --- |
| `MISSING MODEL` | No usable chat model is selected. | Configure a compatible model. Select it. |
| `MISSING PROVIDER` | The selected model has no usable provider. | Repair provider configuration, credentials, or local server details. |
| `Idle` or `Ready` | The composer is available for a request. | Check the context. Send a small prompt. |
| `Typing` or `Generating...` | A request is in progress. | Wait for completion or an error. |
| `Error` | Generation failed. | Read the error. Correct its cause. Retry the same bounded prompt once. |
- **Current** means that the model is selected. Run the model test to check authorization and provider access.
- **Test** must succeed when the model setup provides that action.
- Choosing a model changes the default for future responses.
- A completed response retains the model and provider used for its request.
#### Does changing the status-bar model change an earlier response?
No. It changes the default for later responses. Completed responses retain the model metadata from their own request.
### Use the thread toolbar
The toolbar contains:
- **New Chat**
- **Chat History**
- **Chat Settings**
- **Chat Help**
- a thread-name field
A changed thread name is saved when the field is committed with Enter or loses focus.
### Write and send a message
The composer accepts normal text and large pasted input. Its context hint is `Use @ to add context. eg Based on my notes`.
- Type `@` to choose context for the current response.
- **Send** remains unavailable when the composer is empty or no usable model is configured.
- **Send shortcut** selects what sends from the chat input. The default is `Shift + Enter`; when the selected shortcut is not pressed, Enter inserts a line break.
- **Stream responses** controls whether partial responses appear as they are generated. A partial stream is not a completed response.
- `Typing` or `Generating...` means the request is in progress.
- `Error` means that generation failed. It is not a completed response.
- After correcting the reported cause, retry the same bounded prompt before changing several variables.
### Add known notes as context
**Add context** opens Smart Context for the current response. After context is attached, the control changes to **Open context builder**.
*Select the exact note, then review the resulting one-source tree before Send.*
Each response keeps its own reviewed context set. A later request includes prior non-excluded conversation messages, but it does not automatically copy the prior response's attached source context. Reopen the builder or select the required sources again.
### Find context with Lookup
**Lookup context** uses the current question to retrieve likely notes or blocks.
On the first response, enter at least three words or more than ten characters before expecting **Lookup context** to enable. After a completed response, it can be available for a continuation.
Review the proposed sources. Remove weak, stale, duplicated, or unrelated items before sending. Lookup proposes context. It does not make the sources authoritative.
In **Chat Settings**, **Results limit** bounds the number of candidates proposed by Lookup context. **Result type** selects whole-note **Sources** or smaller **Blocks**. These settings change the proposed review set, not source authority or response correctness.
### Drag notes, results, and named contexts into a thread
Drop any of these onto the active Smart Chat thread to add them to the current response context:
- notes and supported files from File Navigator
- Connections list results or graph nodes
- Lookup results
- named-context dashboard rows
Notes, blocks, and results are added directly. A named context is added as one reusable selection and expands to its saved sources.
#### Add a saved named context from the selector
You can also add a saved context without dragging it. Choose **+ Named contexts** in the context selector. Select one saved context, then expand its source tree. Smart Chat preserves the named origin so the bundle remains distinguishable from unrelated direct attachments.
*Choose one saved context, then inspect every expanded source before Send.*
Review the context tree before sending. If nothing is added, the item can be unsupported, unresolved, or unavailable to the current Smart Environment. Use **Add context** as the fallback.
### Send intentionally without context
When no context is attached, Smart Chat can offer:
- **Lookup context**
- **Select context**
- **Continue without context**
- **Cancel**
Choose **Continue without context** only when a general, ungrounded conversation is intentional. For a note-grounded request, cancel the send. Then attach or retrieve the intended sources.
The dialog appears only after a send attempt with a usable model and no attached context.
#### Why did Smart Chat warn before sending?
The current response had no attached context. Choose Lookup, select known sources, continue intentionally without context, or cancel.
### Include or exclude prior messages
Exclude an earlier exchange when you do not want it to influence later responses. The exchange remains visible in the saved thread. Exclusion changes only the prior-message history sent with future requests.
Keep decisions, constraints, and ground truth. Exclude false starts, discarded drafts, and wrong assumptions.
Use **Include** to restore an excluded exchange to future conversation history. Neither action rewrites the completed response.
### Custom instructions
Open **Chat Settings**. Then choose **Custom instructions** to manage the default instructions added to new threads. Smart Chat also supports instructions saved for one thread.
Thread instructions take precedence in that thread. Review persistent instructions when an otherwise well-grounded response behaves unexpectedly.
### Chat History
Choose **Chat History** from the thread toolbar when you want to reopen a recent saved thread. Suggestions can show the thread name and a recent user-prompt preview.
Type to filter the displayed suggestions. Press `Enter` to open the selected thread. History does not search every word in every message. See [Use Chat History](https://smartconnections.app/docs/chat/#use-chat-history) for deletion shortcuts and search limits.
#### Can Chat History search every word in every message?
No. History searches thread names and displayed recent user-prompt previews, not complete message bodies.
### Chat Manager
Run **Open: Chat Manager view** from the Command Palette under Smart Chat Pro when you need deliberate thread maintenance.
Search by thread name or key. Open uncertain records. Use the row-specific **Rename** or **Delete** action only after the intended thread is visible. See [Manage Smart Chat threads](https://smartconnections.app/docs/chat/#manage-smart-chat-threads-in-obsidian) for bulk selection, counts, confirmation behavior, and recovery.
### Thread data, persistence, and sync
Provider codeblocks store thread URLs in Markdown. Smart Chat API Extension stores its threads, responses, and attached context as Smart Plugin application data rather than ordinary Markdown notes.
That application data can remain device-local unless your sync setup explicitly includes it. During synchronization, do not change the data on several devices at the same time. Reopen one thread. Make sure that it still contains its messages and context. Provider bookmarks stored in Markdown continue to sync with their notes.
### What Smart Chat sends
Smart Chat sends only what the selected workflow needs:
- A provider codeblock sends what you type or upload through that provider's web interface. The active note and vault are not attached automatically.
- Smart Chat API Extension sends the prompt, included prior messages, and the source context selected for that response. It does not send the whole vault automatically.
- A cloud model provider receives that request under the provider's data-handling terms. A local model can keep the request on the machine when its local runtime is active.
Before you send sensitive material:
1. Open the context tree.
2. Remove unrelated sources.
3. Make sure that the correct model or provider is selected.
4. Review the destination's retention settings.
#### Are Smart Chat API Extension threads the same as provider codeblock bookmarks?
No. Provider codeblocks store URLs and active/done fields in Markdown. Smart Chat API Extension threads use Smart Environment records.
### Review context before sending
Review source names, blocks, origin badges, and size estimates in the current Smart Context tree.
Prior non-excluded user and assistant messages can influence a later response. Attached source context is request-specific and does not automatically copy to the next response. Select or reuse the intended source set again. Remove stale items before sending.
### Review completed responses
A completed response can show recorded model information, included context, the user message, and the assistant response. **Include** or **Exclude** affects future conversation history. It does not rewrite completed output.
Before promoting an answer into a trusted note:
1. Make sure that the intended model handled the request.
2. Open the attached context.
3. Examine the source set.
4. Compare factual claims with the cited or attached passages.
5. Separate source-backed statements from suggestions.
6. Treat a fluent but unsupported answer as a bad response, not a successful grounded result.
### Recover from model and completion errors
| Failure | Recovery |
| --- | --- |
| `MISSING MODEL` | Under **Chat models**, choose **+ New**.
Complete its provider and **Chat Model** fields.
Select it under **Default chat model**.
Make sure that the row is **Current**.
Run **Test** or **Test model**. |
| `MISSING PROVIDER` | Repair the provider, credential, or local server configuration. |
| Model test returns `401` | Replace or repair the provider credential before sending note content. |
| Request ends in `Error` | Read the error.
Correct its cause.
Retry the same bounded prompt once. |
| Provider rate limit, exhausted quota, or temporary unavailability | Wait until the provider permits another request, repair billing or quota when applicable, or select another working provider.
Retry one bounded request. |
| Unsupported attached source type or oversized request | Remove the unsupported source, convert it to a supported format, or reduce the request.
Retry the request. |
| Response is generic or contradicts sources | Keep the completed response for comparison.
Reattach the intended context.
Tighten the evidence requirement.
Send a new request. |
| Expected sources are missing on the next response | Open **Add context** or **Open context builder**.
Select the sources again.
Attached context does not automatically carry forward. |
## Manage Smart Chat threads in Obsidian
Smart Chat API Extension stores conversations as thread records. Use **Chat History** to reopen a recent thread quickly and **Chat Manager** for deliberate maintenance.
> [!NOTE] Use Chat Manager for deliberate maintenance
> 1. Search for the target record.
> 2. If its truncated name is uncertain, open the record.
> 3. After a rename, make sure that the new name is saved.
> 4. Use the name as a deletion cue only after this check.
> 5. Delete only after the intended row or selection is visible.
### Open Chat Manager
Run **Open: Chat Manager view** from the Command Palette under Smart Chat Pro.
Chat Manager lists Smart Chat API Extension thread records. It does not list provider URLs stored only in Markdown codeblocks.
#### Is Chat Manager the same as a Dataview Chat Inbox?
No. Chat Manager manages Smart Chat API Extension thread records. A Chat Inbox lists `chat-active` and `chat-done` fields stored in notes.
### Search and filter
Use `Search chats by name or key` to filter visible rows. Filtering does not rename or delete hidden records.
### Open a thread
Use **Open** to load the selected thread in Smart Chat. Open uncertain threads before renaming or deleting them.
### Rename a thread inline
Rename happens inline in the row with **Rename**, **Save**, and **Cancel**.
1. Enter rename mode.
2. Change the name.
3. Save or cancel.
A saved name remains attached to that thread record.
After choosing **Save**, make sure that the changed name remains on the intended row. Then use it as a cleanup cue.
#### Can I rename a thread without opening it?
Yes. Rename is inline in the manager row.
### Delete one thread
Single-thread deletion requires confirmation beside the target row. Verify the thread name before confirming.
This removes the Smart Chat API Extension thread record. It does not remove provider-thread URLs stored in Markdown codeblocks.
### Select and delete visible threads
Use **Select all visible** or select individual visible rows. Then choose **Delete selected**.
1. Filter to a narrow working set.
2. Select only the intended visible rows.
3. Choose **Delete selected**.
4. Confirm the bulk action.
5. Review the remaining list.
#### Does bulk delete remove every thread matching the search?
No. It removes only selected visible rows after confirmation.
### Review thread and message counts
The manager shows thread and completion totals for the current list. Counts support review. They do not indicate that a thread is safe to delete.
### Confirmation and refresh states
Delete confirmation stays inside the manager and beside the affected row or selection. Use **Refresh** after an external change or when the visible list appears stale.
### Use Chat History
**Chat History** is optimized for opening a saved thread from the active Smart Chat workspace. Chat Manager owns deliberate filtering, row controls, and count review.
#### Open History
Choose **Chat History** from the active thread toolbar.
History lists saved, non-deleted threads.
#### Thread suggestions
Each suggestion identifies a saved thread and can include its name plus a recent user-message preview.
#### Completion search
History filters saved thread names and displayed recent user-prompt previews. It does not provide exhaustive search across every word in saved messages.
#### Open the selected thread
Press Enter on a suggestion to open that thread in Smart Chat.
#### Delete from History
On macOS, press `Cmd + Enter` on a History suggestion to request deletion. On other platforms, press `Ctrl + Enter`. Confirm the named thread before deleting it.
##### What does Mod+Enter do in History?
It requests deletion of the selected History thread: `Cmd + Enter` on macOS and `Ctrl + Enter` on other platforms. The confirmation names the thread and states that deletion cannot be undone.
#### Search limitations
Search is designed for fast thread retrieval by the information shown in suggestions. Use Chat Manager for deliberate record maintenance.
##### Why can I find a thread by name but not by a phrase from the middle of the conversation?
History does not perform exhaustive full-message search.
### Chat History vs Chat Manager
Use **Chat History** to reopen or delete a known recent thread quickly. Use Chat Manager to:
- filter names or keys
- examine counts
- rename threads
- manage selected rows
Neither tool searches every word in every message.
### Troubleshooting thread management
- When a thread is missing, clear the filter.
- After you clear the filter, choose **Refresh**.
- Choose **Save** before expecting an inline rename to persist.
- History searches names and displayed recent user-prompt previews.
- Chat Manager searches names and keys.
- Review visible selected rows again after filtering.
- Open an uncertain thread instead of deleting it based only on age.
> [!NOTE] When to use Connect Pro
> Smart Chat does not apply arbitrary AI-generated edits to your notes. It does write its own thread/bookmark state into Markdown and Smart Plugin application data. Use [Connect Pro](https://smartconnections.app/docs/connect-pro/) when you want an explicit general-purpose vault change.
## Related documentation
- [Build reviewable thread context](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian)
- [Find notes by meaning](https://smartconnections.app/docs/lookup/#search-obsidian-notes-by-meaning-with-smart-lookup)
---
## Faq
canonical: https://smartconnections.app/smart-context/faq/
html_url: https://smartconnections.app/smart-context/faq/
markdown_url: https://smartconnections.app/smart-context/faq.md
llms_url: https://smartconnections.app/smart-context/faq/llms.txt
last_modified: 2026-09-06T18:36:21.915Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Context FAQs What is Smart Context for Obsidian?#Smart Context packages selected Obsidian content so you can review exactly what will be copied, exported, or sent.What should I copy first?#Open the note that owns the task. Use the procedure for your installed edition: Core: Run Smart Context: Copy current text to clipboard (choose link depth). In Copy context, select the Depth 0 row marked…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-context/bases/
- title: Builder
url: https://smartconnections.app/smart-context/builder/
- title: Canvas
url: https://smartconnections.app/smart-context/canvas/
- title: Clipboard
url: https://smartconnections.app/smart-context/clipboard/
- title: Codeblock
url: https://smartconnections.app/smart-context/codeblock/
# Smart Context FAQs
## What is Smart Context for Obsidian?
[Smart Context](https://smartconnections.app/smart-context/) packages selected Obsidian content so you can review exactly what will be copied, exported, or sent.
## What should I copy first?
Open the note that owns the task. Use the procedure for your installed edition:
- **Core:** Run `Smart Context: Copy current text to clipboard (choose link depth)`. In **Copy context**, select the **Depth 0** row marked **Current note**. Selecting it copies immediately.
- **Pro:** Select the **Smart Context: Copy current** ribbon action. Open **Depth 0 - current note**. Select **Copy text**.
Paste the result into a clean destination. Examine the result before you send it.
Use Builder only when the active note is not enough.
## What do Depth 0, Depth 1, Depth 2, and Depth 3 mean?
- **Depth 0** starts with the current note and material it embeds. It does not follow ordinary links.
- **Depth 1 - outlinks only** also follows outgoing links.
- **Depth 1 - include backlinks** follows outgoing and incoming links.
- Higher depths repeat that traversal one link layer at a time.
Choose the smallest depth that contains the evidence you intend to send.
## How do I know Smart Context worked?
The pasted content must match the selected depth or reviewed Builder tree. For a model workflow, the answer must use details from those sources. Otherwise, it must identify a specific evidence gap. A copy notice alone does not prove the package is correct.
In this Pro example, the compiled result keeps the source wrapper, includes the reviewed **Template** section, and preserves the following **Review boundary**. For a current-note copy, perform the same check against the depth and source tree you selected.
## Which source modes are current?
Core Builder exposes **Notes**, **Sections**, and **Named contexts**. Pro adds **Media**, **Folders**, **Linked notes**, **Similar notes**, **Tags**, and **External files**.
Each result is a candidate until you deliberately add it. Review the active tree after every addition.
## Which output should I choose?
- **Copy text** for a readable context tree and source bodies in Core or Pro.
- **Copy media** for one composite clipboard image made from selected images and PDF pages in Pro.
- **Copy ZIP** for an archive file reference in Pro.
- **Copy link tree** for a compact source hierarchy. Core exposes a current-note link-tree command. Pro also exposes package menus.
- **Copy with Template** when Smart Templates is installed and the package contains items.
**Copy text** and **Copy media** are separate because some destinations handle text and images differently. When you need both, copy each output separately.
Use **Copy link tree** when the destination needs a compact outline of the source hierarchy instead of the note contents, and when that option is available. Use **Copy text** when the destination needs the contents of the sources.
Paste or open the result. Compare it with the reviewed package.
## When should I use Builder instead of a direct file action?
In Core, use Builder for source-by-source review, precise sections, source and token estimates, and named reuse. Use Pro Builder when you also need dynamic folder or tag groups, Rules, exclusions, media, or external files.
Run the command for your edition:
- Core: `Smart Context: Open new context in builder`
- Pro: `Smart Context Pro: Open new context in builder`
In either edition, you can also select **Smart Context: Open Builder**.
## Can I copy selected notes or folders directly from the file navigator?
Yes. For a direct copy:
1. Select at least two supported notes or folders.
2. Open the context menu.
3. Select **Copy selected notes as context** or **Copy selected folders as context**.
The selection copies directly without a link-depth chooser. Select **Open selection in Context Builder** when you must:
- remove sources
- select sections
- estimate the package size
- save the package for reuse
Run **Select folder to copy contents** when a searchable folder picker is faster.
Direct file and folder actions use the supported content in the files or folders you selected. Use current-note copy when you want Smart Context to follow links from the current note.
## How do Context rules work?
Rules are a Pro feature. They record dynamic package boundaries. An **Include** can contribute a folder or another group. An **Exclude** can keep an exact child or heading out while the group remains included.
What the numbers show:
1. The visible `1 rule` chip preserves continuity with the reviewed package.
2. The global `1 exclude` count and override show the current exclusion state.
3. The Template global heading-exclusion row is visibly On for this package.
4. Source-mode tabs remain available while Rules are inspected.
An Include rule contributes a dynamic source. An Exclude rule keeps matching content out. Turning off a global heading exclusion creates an exception for this context without deleting the global rule. Copy again after any change. Examine the result.
## Can I save reusable context sets?
Yes. Name a reviewed package in Builder. Reopen it through **Named contexts** or the named-context dashboard.
Recheck source availability and size before reuse. In Pro, also check Rules. Saved groups can expand when Pro rules resolve, so the Builder source count can be larger than the dashboard item count.
## How do I review a named context before reuse?
1. Open the dashboard.
2. Make sure that the saved name and item count are correct.
3. Select **Open in context builder** before copying.
**Make a copy**, **Clear this context**, and **Delete context** change the saved definition. Make sure that the selected name is correct before you use these actions.
## What is the difference between a named context and a codeblock?
Use a named context when the same reviewed package should be reused across tasks. Use a Smart Context codeblock when the source manifest should remain visible inside one note.
Simple rule:
> Named context = reusable package. Codeblock = visible manifest attached to a note.
## What can a Smart Context codeblock contain?
Core codeblocks accept local note paths and `ctx::` named-context references. Pro also accepts supported relative external include or exclude lines.
Run the codeblock command for your edition:
- Core: `Smart Context: Insert codeblock`
- Pro: `Smart Context Pro: Insert codeblock`
Compare the fenced source with the rendered tree. After you copy or open the codeblock in Builder, make sure that each path resolves to the intended source.
## Can I send Smart Connections results into Context?
Yes. Select **Send to Smart Context** from a reviewed Connections result set. Make sure that the same sources appear in Builder. If the action is unavailable, open the Smart Plugins Store. Install Smart Context if it is not installed. Enable Smart Context if it is disabled.
## How do I keep a package small enough to review?
- Start from one task-owning note.
- Add a Section when only one heading matters.
- Prefer reviewed Linked notes over broad discovery.
- Remove irrelevant direct sources.
- Inspect Rules for items inherited from folders or tags.
- Treat the token estimate as directional, not provider-exact.
## Can I include only part of a long note?
Yes. Add a **Section**, heading, or block instead of the whole note. Examine the resolved package before copying. This keeps unrelated material out of the package.
## Can I customize the copied format?
Yes. Smart Context settings provide XML, Markdown, JSON, and custom output templates. The context template can insert `{{FILE_TREE}}`. Item templates can use path, name, modified-time, link-depth, extension, and current-note variables. Test the format with a small package. Examine the pasted result before you use the format broadly.
## What if a source is missing or a folder is truncated?
Restore or remove a missing source. If a folder is marked truncated, use one of these actions:
- Narrow the folder.
- Use a **Section**.
- Split the package.
Cancel an incomplete-output confirmation unless a partial package is intentional.
## What if Copy media is too large?
In Pro, reduce the number of selected images or PDF pages. Lower the media resolution if necessary. Copy again. Examine the composite image in the destination.
## What if the clipboard result is empty?
Make sure that the package contains at least one included item. In Pro, make sure that Rules did not exclude everything. Keep Builder open. Reduce the package. Retry the copy. In Pro, use ZIP when a clipboard workflow is unsuitable.
## Does Smart Context upload my notes?
Smart Context assembles the package in Obsidian. Content can leave Obsidian when you copy, paste, export, or send it to another application or provider. Inspect the package before sending sensitive material.
See the [Privacy Policy](https://smartconnections.app/legal/privacy-policy/).
## Can I use Smart Context with ChatGPT, Claude, Gemini, or Smart Chat?
Yes. Smart Context prepares a reviewed package for another destination. Check the included sources. Then paste, upload, or attach the package. The destination receives the content you send under its own privacy and retention terms.
## Does Smart Context change my notes?
Building or copying a package does not rewrite source notes. Saving a named context stores the package definition. A codeblock changes a note only when you choose to insert or edit that manifest. Rules change the package boundary, not the source files.
## Can I include images and PDFs?
Yes, with Pro. Add the intended images or PDF pages through **Media**. Select **Copy media**. Smart Context places one composite image on the clipboard. Make sure that every intended image and page appears in the destination.
**Copy media** includes only the images and PDF pages it finds. Run **Copy text** separately when the task also needs note text, paths, or other written source material.
*The chooser reports the media found at the selected depth before copying.*
## What does the Copy media count tell me?
It shows how many media items Smart Context found in the current note, selection, or context. An unexpectedly small count usually means that the selected depth did not reach every attachment.
## Why did an image or PDF not appear in Copy media?
An image embedded in the starting note can be available at Depth 0. An attachment reached through a link requires enough depth to reach it. Check the media count, then retry in an empty composer that accepts pasted images. Use **Copy text** when the destination is text-only.
## Can I include repos or external files?
Yes, with Pro on desktop. Use **External files** or supported relative-path codeblock syntax. Prefer portable paths such as `../example-project/src`. Explicitly exclude generated output or tests.
Expand the resolved tree. Make sure that included files are present. Make sure that excluded folders or patterns are absent before copying.
## Can I include Canvas or Bases?
Yes. Core and Pro Notes mode searches Markdown notes, Bases, and Canvas files. Pro adds richer Bases rendering and advanced workflows. Add the source. Then examine the resolved tree.
After copying, inspect which Canvas nodes, embedded cards, Base rows, filters, properties, and rendered text reached the destination. Do not infer the compiled output from the source view alone.
Copying text from a Canvas includes supported content from its file cards. It does not preserve card positions, groups, or edge meaning. Use **Copy media** or another visual output when the arrangement itself matters.
## Why did a Base copy more rows or records than I expected?
The rows visible in the Base are not always the exact copy boundary. Builder shows the resolved tree. Depth 0 uses the rendered Base; Depth 1 follows known linked notes. Filters, properties, links, rendered text, and relative `this.file` or `this.note` references affect the compiled output.
## What if a Base does not appear as a Context source?
Make sure that the `.base` file exists, then try adding it through **Builder > Notes**. For version-specific troubleshooting, use the current [Context documentation](https://smartconnections.app/docs/context/#use-bases-as-a-source), because older Environment or exclusion labels may no longer match the interface.
## What should I do when a Base omits a required reference?
Check that the Base displays the link or record you need. Widen to Depth 1 only when a known linked source is missing. When you need exact control, add the required note directly in Builder instead of increasing depth again.
## What does Context Pro add over Core?
Core covers the first useful workflow: current-note copy with a depth chooser. It also supports direct selected-note and folder copy, Builder note and block selection, named contexts, note and named-context codeblocks, template-driven exports, and Canvas sources.
Pro adds:
- images, PDF pages, and other Media sources
- external files, folders, and repositories
- dynamic folder and tag groups in Builder
- exclusions and heading filters
- richer Bases rendering and advanced workflows
Direct folder copy is Core. Dynamic folder and tag groups in Builder are Pro. Basic `.base` source handling is not a Pro-only claim.
## Does Smart Context work on mobile?
Yes for supported in-vault notes, sections, named contexts, codeblocks, and text copy. External file and folder browsing requires desktop filesystem access, and external items in a desktop-created named context cannot resolve on mobile. Clipboard limits vary by device and destination. Begin with a small package. Examine the pasted result before you send it.
---
## Faq
canonical: https://smartconnections.app/smart-plugins/faq/
html_url: https://smartconnections.app/smart-plugins/faq/
markdown_url: https://smartconnections.app/smart-plugins/faq.md
llms_url: https://smartconnections.app/smart-plugins/faq/llms.txt
last_modified: 2026-09-06T17:52:25.774Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Plugins FAQs What are Smart Plugins?#Smart Plugins add focused Obsidian workflows for finding notes, preparing context, reusing prompts, managing chats, exploring graphs, cleaning up duplicates, and running approved vault actions.Which Smart Plugin should I start with?#Start from the material already in front of you: an open note -> Connections a question -> Lookup exact words or a filename…
suggested_links:
- title: Connections Early Getting Started
url: https://smartconnections.app/smart-plugins/connections-early-getting-started/
- title: Core Vs Pro
url: https://smartconnections.app/smart-plugins/core-vs-pro/
- title: For Entrepreneur Coders
url: https://smartconnections.app/smart-plugins/for-entrepreneur-coders/
- title: Getting Started
url: https://smartconnections.app/smart-plugins/getting-started/
- title: Mobile
url: https://smartconnections.app/smart-plugins/mobile/
# Smart Plugins FAQs
## What are Smart Plugins?
[Smart Plugins](https://smartconnections.app/smart-plugins/) add focused Obsidian workflows for finding notes, preparing context, reusing prompts, managing chats, exploring graphs, cleaning up duplicates, and running approved vault actions.
## Which Smart Plugin should I start with?
Start from the material already in front of you:
- an open note -> Connections
- a question -> Lookup
- exact words or a filename -> Obsidian search
- a topic to map -> Graph
- sources to review and reuse -> Context
- a reusable prompt structure -> Templates
- an AI conversation to keep with a note -> Chat
- a reviewed vault action for chat to run -> Connect Pro
Use [Getting Started with Smart Plugins](https://smartconnections.app/smart-plugins/getting-started/) for the first workflow.
## Does Installed mean a Smart Plugin is active and working?
No. Store, Environment, and product-result states are separate.
| Store label or action | Meaning |
| --- | --- |
| **Installed** + **Enable** | Files are installed, but the plugin is disabled. Select **Enable**. |
| **Enabled** | The plugin is enabled in Obsidian but is not reported as loaded and active in the current session. |
| **Active** | The plugin is loaded. |
| **Update to vX** | An eligible newer package is available. |
| **Reload required to activate vX** | Reload Obsidian so the installed version becomes active. |
| **Reload required to disable** | Reload Obsidian to finish disabling it. |
| **Reload required for Smart Environment** | Reload Obsidian so the required Environment state can take effect. |
The Store does not show a standalone **Reload** state or a visible **Configured** state. After **Active**, complete any model, provider, or plugin settings. When the workflow uses indexed sources, confirm that Smart Environment is **Ready**. Then test one result.
Use **Core** and **Pro** to identify the product track. Use **Active**, **Installed**, **Open settings**, and **Enable** on the live row to identify the next lifecycle step.
## What do the other Store labels mean?
- **Install** and **Install Core** install an eligible Core package.
- **Install Pro** installs the Pro package when the account has access.
- **Requires Pro** means current access does not permit the Pro install.
- **Included in Pro** identifies a capability included in the Pro track.
- **Core installed** identifies the installed Core package in a grouped Core/Pro row.
- **Open settings** is available for an active plugin when settings apply.
## How do I recover the Store account row?
Follow the state shown:
- **Checking session...** -> wait for the check to finish.
- **Connect account** -> select **Login**, or **Copy link to login instead**.
- **Session needs refresh** -> select **Refresh**, or **Logout** to remove the saved session.
- **All-access subscription expired** -> select **Get Pro**, **Update subscription**, or **Refresh**, as appropriate.
Signed-out users can browse and install eligible Core plugins. Pro installation and updates require current Pro access.
## Why does an Active plugin still not work?
**Active** means the plugin is loaded. Configuration and a successful product result are separate.
1. Complete its model, provider, or plugin settings.
2. If the workflow uses indexed sources and Environment says **Smart Environment not loaded** or **Idle**, select **Load Smart Environment**.
3. For an indexed-source workflow, select **Resume embedding** when embedding is paused or **Run re-import** when re-import work is queued.
4. For an indexed-source workflow, continue when the status view shows **Smart Environment ready** or **Ready**.
5. If the workflow uses a provider codeblock or Smart Chat API Extension without **Lookup context**, skip the Environment steps. Confirm the provider or model instead.
6. Test one inspectable plugin result.
If one note is missing, use **Inspect active note**. If several notes are missing, use **Show stats** or **Environment stats**. Review eligibility, current embeddings, and exclusions.
## Is deferred loading on mobile a bug?
No. Smart Environment can defer loading on mobile. Select **Load Smart Environment**. Keep the status view open until it shows **Smart Environment ready** or **Ready**. Test the first plugin result separately.
| Workflow | Mobile boundary |
| --- | --- |
| Connections and Lookup | Wait for Environment **Ready**, then make sure that one actual result works. |
| Context using in-vault notes, blocks, and named contexts | These workflows can remain useful. Examine clipboard behavior in the destination app. |
| External files and folders | Desktop filesystem access is necessary. |
| Provider chat codeblocks | Saved links can remain useful. Embedded webviews and automatic URL capture are desktop-dependent. |
| Smart Chat API Extension | Its Smart Plugin application-data storage differs from Markdown provider bookmarks. |
| Connect Pro | Local MCP is same-device only. Any separate cross-device ChatGPT route requires its own currently supported setup and readiness check, with the desktop vault running. |
## Do I need to complete every milestone?
No. [Smart Milestones](https://smartconnections.app/docs/plugins/#track-smart-plugin-progress-with-smart-milestones) records supported first-use actions. Complete only the groups that match your workflow.
What the numbers identify:
1. Overall detected progress.
2. A completed Environment group.
3. A partially completed Connections group.
4. A partially completed Lookup group.
A checked row means its action, event, or installed-plugin state was detected. Judge result quality in the plugin itself.
## How should I evaluate Pro?
Choose one Pro workflow with a clear job in your vault. Complete its prerequisites. Produce one real result. Decide whether that result improves the work before you add another Pro workflow.
---
## Faq
canonical: https://smartconnections.app/pro/faq/
html_url: https://smartconnections.app/pro/faq/
markdown_url: https://smartconnections.app/pro/faq.md
llms_url: https://smartconnections.app/pro/faq/llms.txt
last_modified: 2026-09-06T17:51:59.680Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Pro Plugins FAQs What are Pro plugins?#Pro plugins add advanced Smart Plugin workflows. Examples include configurable retrieval, reusable external or media context, Smart Chat API Extension, Smart Graph, large-vault indexing, and supported vault actions through Connect Pro. Start with the one workflow that removes a real bottleneck. You do not need to install or enable every Pro plugin.What if I…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
# Pro Plugins FAQs
## What are Pro plugins?
[Pro plugins](https://smartconnections.app/pro-plugins/) add advanced Smart Plugin workflows. Examples include configurable retrieval, reusable external or media context, Smart Chat API Extension, Smart Graph, large-vault indexing, and supported vault actions through Connect Pro.
Start with the one workflow that removes a real bottleneck. You do not need to install or enable every Pro plugin.
## What if I am already a supporter?
Supporters from before 2026 are grandfathered into All-access. Sign in with your existing supporter key. If access is missing, reply to the original welcome email or use the support path on the [Pro page](https://smartconnections.app/pro-plugins/).
## Are free Core plugins going away?
No. Core remains free and source-available under the [Smart Plugins License](https://smartconnections.app/legal/license/). That license is not an OSI open-source license.
## Does Pro upload my notes?
Pro access does not upload notes by itself. Smart Plugins are local-first by default.
A configured cloud embedding or ranking provider can receive eligible source text needed for that operation. Provider-backed chat and other downstream workflows can also receive the prompt or context you explicitly send through them. Review the context package and provider before sending a Smart Chat message or using another verified remote-model action. See the [Privacy Policy](https://smartconnections.app/legal/privacy-policy/).
## Where are Pro API keys stored?
Supported Pro integrations can store API keys through Obsidian's native keychain-backed secret storage. This protects the credential. It does not change which reviewed note content is sent when a provider-backed action runs. Examine the selected context and provider before you send sensitive material.
## Is Pro complicated to set up?
Setup depends on the plugin. Use this sequence:
```text
Installed -> Enabled -> Active -> Configured or Ready -> Successful result
```
An **Active** Store row means the plugin loaded in the current Obsidian session. It does not mean that a required model, provider, source index, or connection is ready, and it does not replace a visible product result.
The [Pro getting-started guide](https://smartconnections.app/pro-plugins/) walks through one complete result before introducing the other plugins.
## What if I am not sure Pro is for me yet?
Use the trial shown on the [Pro page](https://smartconnections.app/pro-plugins/) to test one real workflow. The current offer includes a 14-day All-access trial. Confirm the current price, renewal date, and cancellation terms at checkout.
Choose one bottleneck. Complete only its prerequisite. Judge the result in its destination. Avoid changing several plugins or shared settings at once.
## What is included in the 14-day free trial?
The current offer provides 14 days of All-access with no charge on the first day. Use the live Pro page to see the packages included now. Use the billing portal to see the renewal date and terms attached to your account.
## What does Experimental mean in the Store?
Experimental plugins are available to eligible Pro accounts but may change more quickly than the main catalog. Use the live Store grouping as the current authority. Do not make an experimental workflow critical until you test its behavior and recovery path with your vault.
## Do you offer a student discount?
The current Pro offer provides a manual student-discount review. Start the trial. Reply to the welcome email. Follow the current proof-of-enrollment instructions. Send only the information requested.
## What's the process to cancel a subscription?
Use [Manage subscription](https://smartconnections.app/subscription-manage/). Confirm the effective cancellation date and remaining access in the billing portal.
## What if I have not received my key after starting a trial?
Check the email address used at checkout and its spam folder. The welcome email contains the license key and subscription-management link. If it is still missing, contact support through the [Pro page](https://smartconnections.app/pro-plugins/) with non-sensitive purchase details.
## What if I encounter an issue with a Pro plugin?
Include:
- the plugin name
- the exact action you took
- what you expected
- what happened instead
- the complete error message
- your Obsidian version
- Environment readiness when the workflow uses indexed sources
Remove note content, account details, license keys, provider credentials, and local paths from screenshots before sharing them.
## Do I need all Pro plugins to get value?
No. Enable only the workflow that solves the current problem. Milestones and other learning aids are optional.
## Can I switch which Pro plugins I use during the trial?
Yes. All-access lets you test the Pro plugins available to your account in the live Store. Finish one result before switching so you can identify what caused an improvement or failure.
## Will Pro help if my large vault feels slow?
Some Pro workflows include a faster local performance index for Connections, Lookup, Inline Connections, Smart Graph, and Dedupe. Environment stats show index readiness and coverage, not user-perceived speed.
To evaluate a performance change, repeat the same representative workflow on the same vault snapshot and machine. Compare several runs rather than relying on one opening or query.
## What is Connect Pro?
[Connect Pro](https://smartconnections.app/docs/connect-pro/) links the Official Connect Pro GPT in ChatGPT to a running Obsidian Desktop vault. The connection uses Connect Pro and Obsidian CLI.
It is a connected command session, not Obsidian Sync, a mobile-vault replacement, or a hidden always-on agent.
## Does Connect Pro require anything extra?
You need:
- Obsidian Desktop running with the intended vault open
- Obsidian 1.12.2 or later
- Obsidian CLI enabled
- Connect Pro installed and enabled
- for the remote Official GPT path: Connect Pro connected and the Official Connect Pro GPT available in ChatGPT
- for Local MCP: the Local-only MCP server Running and an approved local MCP client; the remote GPT tunnel is not required
For the remote Official Connect Pro GPT route, continue when Connect Pro shows Connected with **Disconnect**. For Local MCP, continue when its server is Running and the intended trusted client can complete the bounded read; Remote GPT may remain Disconnected. The [Connect Pro getting-started guide](https://smartconnections.app/connect-pro/getting-started/) begins with one reversible test note.
## How do you decide what goes into Core vs Pro?
Core focuses on broadly useful workflows with little setup. Pro adds deeper configuration, specialized integrations, large-vault infrastructure, and higher-maintenance workflows. See [Core vs Pro](https://smartconnections.app/smart-plugins/core-vs-pro/) for the current comparison.
---
## Getting Started
canonical: https://smartconnections.app/smart-chat/getting-started/
html_url: https://smartconnections.app/smart-chat/getting-started/
markdown_url: https://smartconnections.app/smart-chat/getting-started.md
llms_url: https://smartconnections.app/smart-chat/getting-started/llms.txt
last_modified: 2026-09-04T20:58:30.269Z
usage_notes: |-
Use this page to answer questions about Getting Started.
excerpt: |-
Getting Started with Smart Chat Use a Smart Chat provider codeblock in the note where the conversation belongs. The block can show ChatGPT, Claude, Gemini, or another supported provider. First useful result Send one concrete prompt and receive one visible provider response. Opening the provider, inserting a block, or saving a bookmark is not yet a successful chat. This page starts with Core…
suggested_links:
- title: Api
url: https://smartconnections.app/smart-chat/api/
- title: Codeblock
url: https://smartconnections.app/smart-chat/codeblock/
- title: Faq
url: https://smartconnections.app/smart-chat/faq/
- title: Settings
url: https://smartconnections.app/smart-chat/settings/
- title: Thread Dashboard
url: https://smartconnections.app/smart-chat/thread-dashboard/
# Getting Started with Smart Chat
Use a Smart Chat provider codeblock in the note where the conversation belongs. The block can show ChatGPT, Claude, Gemini, or another supported provider.
> [!NOTE] First useful result
> Send one concrete prompt and receive one visible provider response. Opening the provider, inserting a block, or saving a bookmark is not yet a successful chat.
This page starts with Core provider codeblocks. They use provider web interfaces and do not require an API model or Smart Environment retrieval.
## Choose the right Chat surface
| Surface | Use it when | First-win signal |
| --- | --- | --- |
| Core provider codeblock | One note needs one provider such as ChatGPT, Claude, or Gemini. | The provider returns a visible response. |
| Universal Smart Chat codeblock (Pro) | One note needs a shared `smart-chat` codeblock for supported provider web threads or Smart Chat API Extension thread references. | The selected provider returns a visible response, or the selected API model completes the intended response. |
| Smart Chat API Extension | You want a configured local or cloud model to answer in the Smart Chat workspace, optionally from approved notes. | A completed response appears with the intended model and context. |
For Smart Chat API Extension, use [Smart Chat API Extension Getting Started](https://smartconnections.app/smart-chat/api/getting-started/).
## What you will do
1. Confirm that Core Smart Chat is installed and enabled.
2. Open the note where the conversation belongs.
3. Insert one provider codeblock.
4. Open its provider surface.
5. Send a concrete prompt.
6. Confirm that a response is visible.
7. If the provider creates a recognized durable URL, make sure that Smart Chat saves it.
8. If Smart Chat saves the URL, reopen it.
9. Make sure that the saved URL opens the same conversation.
## Before you start
You need:
- Core Smart Chat installed and enabled
- one note where the chat thread belongs
- Obsidian desktop for the full embedded chat experience
- provider sign-in when that provider requires it
Core provider codeblocks use provider web interfaces. You do not need an API key, a configured chat model, or an Environment index for this workflow.
> [!TIP]
> You can still use local models in Core with the Open WebUI codeblock if you already run Open WebUI locally.
### Activate Core Smart Chat
Use this sequence:
1. Open **Browse Smart Plugins**.
2. Find Chat.
3. Follow the visible **Install** or **Enable** action.
4. If **Enable** appears after installation, choose **Enable**.
5. If the row shows a full **Reload required...** action, complete it.
6. Open a provider codeblock.
7. Send a prompt.
8. Make sure that the provider returns a visible response.
## 1. Open the note where the chat thread belongs
Start from the note that owns the work.
Good candidates:
- project hub
- meeting note
- research note
- decision note
- bug report
- draft
- outcome note
The simple rule:
> Put the chat thread in the note where future-you would look for it.
## 2. Insert a provider codeblock
Open the Command Palette in the owning note. Run the provider action that you need. For ChatGPT, run **Insert OpenAI ChatGPT codeblock**.
Other Core actions include:
- **Insert Anthropic Claude codeblock**
- **Insert Google Gemini codeblock**
- **Insert DeepSeek codeblock**
- **Insert Perplexity codeblock**
- **Insert Grok codeblock**
- **Insert Google AI Studio codeblock**
- **Insert Open WebUI codeblock**
- **Insert Kimi codeblock**
The ChatGPT action inserts:
````md
```smart-chatgpt
```
````
Switch the note to Reading view so the provider surface can render. For fences, providers, and Pro differences, see [Supported providers](https://smartconnections.app/docs/chat/#supported-providers).
## 3. Open the provider surface
In Reading view, open the provider from the rendered block. If the provider requires an account, sign in.
At this point, distinguish these states:
| State | Meaning | Next action |
| --- | --- | --- |
| Provider did not load | The web surface is unavailable or stalled. | Use **Refresh**. If sign-in is required, use Obsidian Web Viewer. |
| Provider loaded | The interface is visible, but no request has succeeded. | Enter the test prompt below. |
| Response visible | The provider returned an answer. | The first Chat win is complete. |
| Durable URL saved | Smart Chat recognized a provider thread URL and wrote it to Markdown. | Reopen it to verify continuity. |
*New chat and Unsaved identify a loaded provider surface, not a completed or durable thread.*
## 4. Send the first prompt
Enter a small, useful prompt:
```prompt
Create a five-item review checklist for this task: [describe the task in one sentence].
Keep each item specific and testable.
```
Replace the bracketed text with the task owned by the note.
Send it with the provider's normal send control.
You know Chat is working when the provider displays a response that fits the request. Use that response for the note's work. When you need more information, send one bounded follow-up question.
> [!IMPORTANT] Response success and thread saving are different
> A provider can return a response while the block still shows a new or unsaved chat. Do not claim durable continuity yet. Wait until Smart Chat writes a recognized provider thread URL to the block. Reopen that URL. Make sure that it opens the same conversation.
## 5. Save the thread when a durable URL exists
After the provider creates a recognized conversation URL, Smart Chat can write the reference into the codeblock. Reopen the saved entry. Make sure that it returns to the intended conversation.
Example saved state:
````md
```smart-chatgpt
chat-active:: 1700000100 https://chatgpt.com/c/example-active
```
````
If the surface still says new chat or no URL appears, the response can still be valid. However, the thread is not yet a durable note bookmark. Continue the provider conversation until it creates a recognized URL. Alternatively, use **Copy link** only when the current URL identifies the conversation.
Core stores saved URLs and Active or Done state in Markdown. It does not store the full provider transcript by default.
## Optional: Mark active or done after review
Use Active or Done only after a saved thread reopens correctly.
| State | Meaning | Use it when |
| --- | --- | --- |
| `chat-active::` | The thread is still relevant, waiting, or in progress. | You need to return later. |
| `chat-done::` | You reviewed the thread and closed the loop for now. | The useful output has been handled, promoted, saved, or dismissed. |
Done does not mean Smart Chat judged the answer correct.
Done means you reviewed the thread and decided the conversation is closed for now.
Keep the thread Active until its backing conversation reopens. Review the result. Change the thread to Done only after this review. Provider completion alone does not determine the state. See [Mark done and Mark active](https://smartconnections.app/docs/chat/#mark-done-and-mark-active).
## What gets saved
A provider codeblock can hold saved Active or Done links. The provider URL and user-owned state stay in Markdown, while the full provider transcript remains with the provider. See [Thread storage grammar](https://smartconnections.app/docs/chat/#thread-storage-grammar).
## When to use the universal Smart Chat codeblock
Smart Chat Pro adds the universal `smart-chat` codeblock. Use this sequence:
1. Run **Insert Smart Chat codeblock**.
2. Switch the note to Reading view.
3. Choose a provider from **New chat**.
4. Send a concrete prompt.
5. Confirm that the provider returns a settled visible response.
*Choose the provider for the new thread.*
*The visible response completes the provider-response check. A recognized provider conversation URL is still required before the thread becomes a durable note bookmark.*
Use it when one codeblock needs a shared provider selector or several saved provider threads. Durable URL capture and Active or Done tracking remain separate checks.
## Desktop vs mobile
| Platform | What to expect | Best use |
| ---------------- | ------------------------------------------------------------------------------------------------ | ---------------------------------------- |
| Obsidian Desktop | Embedded provider surface, sending, URL capture, and thread controls | Full workflow |
| Obsidian Mobile | Saved URLs can remain useful as external bookmarks. The embedded provider workflow is unavailable. | Reopen saved conversations |
| Any text editor | Saved provider URLs and Active or Done fields remain readable Markdown | Recover known links |
On mobile, use saved URLs as bookmarks or external links. Use desktop for the full embedded provider interface. See [the platform reference](https://smartconnections.app/docs/chat/#desktop-and-mobile-behavior).
## Recover the first workflow
| Problem | Recovery |
| --- | --- |
| Provider surface is blank or stalled | Use **Refresh**. If sign-in is required, complete it in Obsidian Web Viewer. Return to the block. Use **Refresh** again. |
| The provider rejects the prompt | Make sure that sign-in works. Make sure that the account has provider access. Make sure that the provider is available. Retry the same small prompt. |
| A response appears but the block still says new chat | Treat response success as complete, but not durable saving. Wait for a recognized conversation URL before relying on the bookmark. |
| A saved entry opens the wrong conversation | Return to the correct provider thread. Save or select its exact URL. Keep the record Active until continuity is confirmed. |
| The embedded provider is unavailable on mobile | Open an already saved URL externally or continue on desktop. |
## Continue only when the first response needs more
| Next need | Continue with |
| --- | --- |
| Ground a provider request with a small reviewed source set | [Smart Context Getting Started](https://smartconnections.app/smart-context/getting-started/) -> [Docs](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian) |
| Find a source by meaning when you know the question, not the note | [Smart Lookup Getting Started](https://smartconnections.app/smart-lookup/getting-started/) |
| Apply a specific vault change | [Connect Pro Getting Started](https://smartconnections.app/connect-pro/getting-started/) |
| Review Active and Done links across notes | [Build a Chat Inbox](https://smartconnections.app/docs/chat/#build-a-thread-dashboard-with-dataview) |
| Use a configured local or cloud API model inside Obsidian | [Smart Chat API Extension Getting Started](https://smartconnections.app/smart-chat/api/getting-started/) -> [Docs](https://smartconnections.app/docs/chat/#chat-with-obsidian-notes-using-local-and-api-models) |
| Understand every provider, selector, action, and recovery path | [Smart Chat codeblock documentation](https://smartconnections.app/docs/chat/#embed-and-organize-ai-chat-threads-in-obsidian) |
## First useful result checklist
Before moving on, check:
| Check | You know it worked when... |
| --- | --- |
| Note chosen | The provider codeblock is in the note where the conversation belongs. |
| Block rendered | Reading view shows the intended provider surface. |
| Prompt sent | The provider accepted the concrete prompt. |
| Response completed | A visible response matches the request. This is the required first win. |
| Durability understood | A saved URL is trusted only after it reopens the same conversation. |
| Next step clear | You know whether to save the provider URL, use the universal Smart Chat codeblock, or move to Smart Chat API Extension. |
## Related pages
- [Smart Chat documentation](https://smartconnections.app/docs/chat/)
- [Smart Context Getting Started](https://smartconnections.app/smart-context/getting-started/)
- [Smart Chat API Extension Getting Started](https://smartconnections.app/smart-chat/api/getting-started/)
- [Smart Connections Getting Started](https://smartconnections.app/smart-connections/getting-started/)
---
## Faq
canonical: https://smartconnections.app/smart-dedupe/faq/
html_url: https://smartconnections.app/smart-dedupe/faq/
markdown_url: https://smartconnections.app/smart-dedupe/faq.md
llms_url: https://smartconnections.app/smart-dedupe/faq/llms.txt
last_modified: 2026-09-04T20:58:25.447Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Dedupe FAQs What is Smart Dedupe?#Smart Dedupe finds likely exact and near-duplicate blocks across notes. It places candidates side by side so you can decide whether to keep them separate, consolidate useful material manually, archive a redundant source, or ignore the match.Does 1.00 similarity mean an exact duplicate?#It is a very strong signal, and an Exact hash match identifies identical…
suggested_links:
- title: Getting Started
url: https://smartconnections.app/smart-dedupe/getting-started/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
# Smart Dedupe FAQs
## What is Smart Dedupe?
[Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) finds likely exact and near-duplicate blocks across notes. It places candidates side by side so you can decide whether to keep them separate, consolidate useful material manually, archive a redundant source, or ignore the match.
## Does 1.00 similarity mean an exact duplicate?
It is a very strong signal, and an **Exact hash match** identifies identical normalized text, but neither tells you what the notes are for. Open both sources before merging, archiving, or deleting anything.
## Can I cancel a scan?
Yes. Choose **Cancel** while the scan is running. Wait for the Detector to settle before you start another scan. Treat any displayed candidates as a partial result set.
*While a scan runs, you can monitor processed-note progress or cancel it. A cancelled scan keeps only the candidates found so far, so treat that result set as partial.*
## What is a good starting threshold?
For a conservative first review, start around `0.90`. Lower the threshold toward `0.85` when you intentionally want to find paraphrases. Raise it when weak matches create noise. The threshold controls which semantic candidates qualify, not whether two passages should be consolidated.
## Does the running scan show how many results have been found?
The Detector shows scan progress, processed-note counts, and **Cancel**, but not a running match count. Candidate rows are available after completion or cancellation.
## Will this delete or merge my notes?
No. Smart Dedupe is review-first. **Copy** and **Open** help you inspect a candidate, but cleanup remains a deliberate action in your normal Obsidian workflow.
## How does it find duplicates?
Smart Dedupe can use an exact-text pass and semantic similarity over prepared block embeddings. Exact matching can work without embeddings. Semantic matching requires Smart Environment to prepare the relevant blocks.
## What if two similar notes are both useful?
Keep them separate. Similar passages can serve different audiences, projects, stages, or decisions. Use [Smart Connections](https://smartconnections.app/docs/connections/) when the relationship is useful. Use Dedupe only when the overlap creates rework, conflict, or context bloat.
## Is Smart Dedupe just another Connections view?
No. Connections is for discovering related notes from the current note. Dedupe is for reviewing repeated work and deciding whether anything should change.
## Will cleanup improve AI output?
It can make context easier to inspect by reducing repeated or conflicting source material, but it does not guarantee a better model answer. After cleanup, rebuild the affected package with [Smart Context](https://smartconnections.app/docs/context/).
## Will a full-vault scan be slow?
It can be heavier than a current-note scan. Start with **Current note**. Use a stricter threshold. Set a bounded result count. Use **Full vault** only after you understand the controls and have a review session you can finish.
## Do Source and Duplicate tell me which note to keep?
No. They name the two sides of a candidate pair. Neither label tells you which note is the original or which passage you should keep.
## Do zero results mean the vault has no other duplicates?
No. Zero results only means that no candidates matched the settings used for that run. Scope, threshold, minimum length, exclusions, result limit, embedding readiness, and cancellation can all change what appears. Change one setting at a time before drawing a broader conclusion.
## When should I use Lookup or Obsidian Search instead?
Use Smart Lookup when you remember an idea or can phrase a question. Use Obsidian Search when exact text, titles, tags, syntax, or regex matter. Use Dedupe when you need to decide what repeated material to keep, combine, archive, or ignore.
## What scope should I start with?
Start with one current note where repeated material already creates a decision. Review one candidate pair to completion before broadening the scan.
---
## Dedupe
canonical: https://smartconnections.app/docs/dedupe/
html_url: https://smartconnections.app/docs/dedupe/
markdown_url: https://smartconnections.app/docs/dedupe.md
llms_url: https://smartconnections.app/docs/dedupe/llms.txt
last_modified: 2026-09-04T20:57:53.334Z
usage_notes: |-
Use this page to answer questions about Dedupe.
excerpt: |-
Smart Dedupe Smart Dedupe finds exact and semantically similar blocks so you can compare both sources before deciding what to keep or change. New to Smart Dedupe? Start with Getting started with Smart Dedupe for one controlled current-note scan. The persistent Duplicate Detector keeps the scan boundary, progress, and candidates in one view. Similarity creates a review question Smart Dedupe does…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Graph
url: https://smartconnections.app/docs/graph/
# Smart Dedupe
Smart Dedupe finds exact and semantically similar blocks so you can compare both sources before deciding what to keep or change.
> [!TIP] New to Smart Dedupe?
> Start with [Getting started with Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) for one controlled current-note scan.
The persistent Duplicate Detector keeps the scan boundary, progress, and candidates in one view.
> [!IMPORTANT] Similarity creates a review question
> Smart Dedupe does not merge, archive, or delete notes. A high score or exact match still requires you to inspect both sources and decide what each should do.
## Find duplicate and near-duplicate notes in Obsidian
Use Dedupe when repeated passages create conflicting drafts, repeated decisions, or bloated AI context. Use Connections when similar notes are useful neighbors rather than cleanup candidates.
Use Smart Lookup when you remember an idea or can phrase a question. Use Obsidian Search when you need an exact text match.
### Prepare the first scan
Smart Dedupe is a Pro plugin. Use this sequence:
1. Enable Smart Dedupe.
2. Open a meaningful note.
3. From the Command Palette, run **Smart Dedupe Pro: Open Duplicate Detector**.
Exact matching can find repeated text without embeddings. Semantic matching requires the relevant blocks to be prepared by Smart Environment. If semantic results are missing, check [Smart Environment readiness](https://smartconnections.app/smart-environment/faq/#how-do-i-know-smart-plugins-are-ready-to-test) before changing the threshold.
### Scan the current note
The current-note scan compares blocks in the active note with candidates in other notes. It excludes pairs from the same note. Start here because the source of every comparison is easy to understand.
1. Open the note that owns the passage you want to review.
2. Choose **Current note**.
3. Set a bounded threshold and result count.
4. Keep exact matches visible for the first pass.
5. Choose **Run note scan**.
### Scan the full vault
The full-vault scan compares blocks across different notes. Use it after a current-note scan has taught you what useful and noisy candidates look like.
Before choosing **Run vault scan**, review the threshold, maximum results, minimum length, and exact/frontmatter options. A full-vault result set is bounded by those choices and may not contain every qualifying pair.
### Open the persistent Duplicate Detector
Run **Smart Dedupe Pro: Open Duplicate Detector**. The current Detector is a persistent Obsidian view, not the older sequence of threshold, progress, and results pop-ups.
Keep the Detector open or return to its tab while a scan runs.
### Configure the scan
Before running a scan, review:
- **Current note** or **Full vault** scope
- **Similarity threshold**
- **Max results**
- **Minimum length**
- **Skip exact matches**
- **Exclude frontmatter matches**
1. **Scan scope** chooses the active note or the full vault.
2. **Similarity threshold** sets the minimum semantic similarity.
3. **Max results** bounds the review set.
4. The checkboxes can hide exact matches or frontmatter matches.
**Minimum length** and the run action appear farther to the right in the full Detector.
### Choose a similarity threshold
Higher values return stricter semantic candidates. Lower values broaden the set and usually increase false positives.
For a conservative first review, start around `0.90`. Lower toward `0.85` when you intentionally want paraphrases or when the stricter pass returns too little. Change one setting at a time so you can tell what improved the result.
### Bound the result count
**Max results** limits the collected review set. Use a number you can realistically inspect in one session.
A full-vault scan may stop after collecting enough qualifying semantic matches. The displayed set is sorted by score, but it is not guaranteed to be an exhaustive global top-N across the vault.
### Exclude frontmatter matches
Enable **Exclude frontmatter matches** when repeated YAML properties or metadata blocks are not useful cleanup candidates.
### Understand exact matching
Exact repeated text is checked separately from semantic similarity. An exact match can appear even when one or both blocks do not have embeddings.
Use **Skip exact matches** only after you have reviewed obvious repeats and intentionally want to concentrate on paraphrases.
### Set the minimum length
**Minimum length** applies to the exact-hash pass only. Raise it when short exact repeats such as headings, labels, or boilerplate dominate. It does not filter semantic candidates; use **Similarity threshold** for weaker semantic matches.
### Monitor progress
During a longer scan, the Detector shows the current phase, a progress bar, the processed and total block counts, `Scanning...`, and **Cancel**. It does not show the scanner's internal `results_found` value. Short scans may finish before the running state remains visible long enough to read.
#### Why might progress be hard to inspect?
A small current-note scan can settle almost immediately. Continue with the completed result instead of rerunning only to watch the progress state.
### Keep or restore the scan view
The current Detector is a persistent view, so it does not use the older pop-up minimize-and-restore workflow. Keep its tab open or return to the Detector while the scan is active.
### Cancel a scan
Choose **Cancel** while a scan is running. Wait for `Scan cancelled.` and a scope summary ending in `| cancelled`. The run action becomes available again after the running state settles. Any candidates left in the view form a partial result set, not a completed scan.
#### Can I cancel a full-vault scan?
Yes. Cancel it in the Detector. Treat the remaining candidates as incomplete.
### Review candidates side by side
Each result shows two block previews with a similarity value and source identity. Check:
1. whether the candidate is exact or semantic
2. the **Source** and **Duplicate** note names
3. the line ranges or headings
4. enough surrounding text to understand each passage's purpose
A score helps order review. It does not tell you which passage is canonical or whether either should change.
### Copy or open a result
Use **Open** when the excerpt is not enough and you need the full note. Use **Copy** when you deliberately need the passage text for comparison. Neither action changes the notes.
Use [Connections](https://smartconnections.app/docs/connections/) for useful related material. After cleanup, rebuild an affected AI package with [Smart Context](https://smartconnections.app/docs/context/).
> [!NOTE] **Source** and **Duplicate** name the two sides of the candidate pair. Neither tells you which note to keep.
### Decide what happens to the overlap
After reading both sources, choose deliberately:
- keep both because they serve different jobs
- rewrite one to make the distinction explicit
- consolidate useful material manually
- archive a genuinely redundant source through your normal workflow
- ignore the candidate
#### Does a score of 1.00 mean the blocks should be merged?
No. It is a strong overlap signal. The notes' roles and surrounding context determine the cleanup decision.
#### Does Smart Dedupe delete or merge notes automatically?
No. It presents candidates for human review.
### Interpret result-set limits
No result set proves that every duplicate has been found. Threshold, scope, minimum length, exclusions, maximum results, embedding readiness, and cancellation all affect what appears.
A settled zero-result scan means no candidates matched the current boundary. It does not mean the vault contains no repeated material.
### Troubleshoot a scan
| Symptom | First adjustment |
| --- | --- |
| No semantic candidates | Confirm block readiness. If the blocks are ready, lower the threshold slightly. |
| Too many weak semantic candidates | Raise the similarity threshold or return to **Current note**. Use **Minimum length** only to suppress short exact-hash repeats. |
| Boilerplate dominates | Enable **Exclude frontmatter matches**. Raise **Minimum length**. |
| Exact repeats hide paraphrases | Review the exact matches. Then try **Skip exact matches** deliberately. |
| Scan stops early | Wait for the status to settle. Treat any candidates as partial. |
| You expected automatic cleanup | Open both sources. Make the change manually. |
## Related documentation
- [Getting started with Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/)
- [Find related rather than duplicate notes](https://smartconnections.app/docs/connections/#find-related-notes-in-obsidian-with-smart-connections)
- [Rebuild a cleaner context package](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian)
- [Check Smart Environment readiness](https://smartconnections.app/smart-environment/faq/#how-do-i-know-smart-plugins-are-ready-to-test)
---
## Getting Started
canonical: https://smartconnections.app/smart-dedupe/getting-started/
html_url: https://smartconnections.app/smart-dedupe/getting-started/
markdown_url: https://smartconnections.app/smart-dedupe/getting-started.md
llms_url: https://smartconnections.app/smart-dedupe/getting-started/llms.txt
last_modified: 2026-09-04T20:57:44.846Z
usage_notes: |-
Use this page to answer questions about Getting Started.
excerpt: |-
Getting started with Smart Dedupe The first useful result is not a cleaner vault. It is one candidate pair you understand well enough to keep, consolidate, archive, or ignore deliberately. Review before changing anything A 100% similarity value or Exact hash match label identifies strong overlap. It does not tell you whether the two passages serve the same purpose. Before you start Confirm that:…
suggested_links:
- title: Faq
url: https://smartconnections.app/smart-dedupe/faq/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
# Getting started with Smart Dedupe
The first useful result is not a cleaner vault. It is one candidate pair you understand well enough to keep, consolidate, archive, or ignore deliberately.
> [!IMPORTANT] Review before changing anything
> A `100%` similarity value or **Exact hash match** label identifies strong overlap. It does not tell you whether the two passages serve the same purpose.
## Before you start
Confirm that:
- Smart Dedupe Pro is installed and active
- the note you want to review is open
- Smart Environment has prepared blocks when you want semantic matches
- the first note contains material you can evaluate safely
Start with a normal working note, not customer data, credentials, legal records, or another high-risk source.
## 1. Open the Duplicate Detector
Open the Command Palette. Run this command:
```obsidian-command
Smart Dedupe Pro: Open Duplicate Detector
```
The persistent **Duplicate Detector** keeps scope, controls, progress, and candidates in one view.
Choose **Current note** for the first pass. Use **Full vault** only after the controls and review process feel predictable.
## 2. Set a conservative first boundary
For the first current-note scan:
1. Keep exact matches visible.
2. Set **Max results** to a number you can review in one session.
3. Use **Minimum length** to suppress short exact-hash repeats such as headings and boilerplate. It does not filter semantic candidates.
4. Exclude frontmatter when repeated YAML is not useful.
5. Start around `0.90` for a strict review. Lower toward `0.85` when you intentionally want paraphrases.
The current controls may open with a different saved value. Review the visible setting instead of assuming it is a universal default.
See [Configure the scan](https://smartconnections.app/docs/dedupe/#smart-dedupe-scan-configure) for every control.
## 3. Run the note scan
Choose **Run note scan**. Let the Detector settle.
A short scan can complete before the running state is easy to read. A longer scan shows progress. It also shows a **Cancel** action. If you cancel, wait for the status to settle. Treat any remaining candidates as a partial set.
## 4. Review one candidate pair
For one promising result, check:
1. exact or semantic match type
2. similarity value
3. **Source** and **Duplicate** note identities
4. headings or line ranges
5. surrounding text in both notes
Choose **Open** when the excerpt does not explain why each passage exists. Use **Copy** only when you deliberately need the text for comparison. **Copy** does not merge anything.
**Source** and **Duplicate** name the two sides of the candidate pair. Neither label tells you which note to keep.
## 5. Make one human decision
Choose the outcome that matches the notes' roles:
- **Keep both** when the passages serve different audiences, projects, or stages.
- **Clarify the distinction** when both notes are useful but too easy to confuse.
- **Consolidate manually** when one source should absorb the useful material.
- **Archive one source** when it is genuinely redundant but you want a reversible cleanup path.
- **Ignore the candidate** when the overlap is harmless.
Smart Dedupe does not perform this decision for you.
## You know it worked when
- the Detector returned at least one understandable candidate, or a clear zero-result state
- you inspected both sources rather than relying on the score
- you made one explicit keep, clarify, consolidate, archive, or ignore decision
- any note change was deliberate and reversible
## If the first scan is weak
| What happened | Try first |
| --- | --- |
| No semantic candidates | Confirm block readiness, then lower the threshold slightly. |
| Too many weak candidates | Raise the threshold, increase minimum length, or reduce Max results. |
| Metadata dominates | Enable **Exclude frontmatter matches**. |
| Exact repeats hide paraphrases | Review the exact repeats. Then try **Skip exact matches**. |
| The scan finished too quickly to watch | Review the settled candidates. Do not rerun only to inspect progress. |
| The result set is empty | Confirm scope and settings. Zero results applies only to that boundary. |
| You expected automatic cleanup | Open both notes. Make the change through your normal Obsidian workflow. |
## Continue from here
- [Understand every Dedupe control](https://smartconnections.app/docs/dedupe/#smart-dedupe-scan-configure)
- [Review candidates and result limits](https://smartconnections.app/docs/dedupe/#smart-dedupe-scan-results)
- [Use Connections for related, non-duplicate material](https://smartconnections.app/docs/connections/)
- [Rebuild reviewed context after cleanup](https://smartconnections.app/docs/context/)
---
## Getting Started
canonical: https://smartconnections.app/smart-chat/api/getting-started/
html_url: https://smartconnections.app/smart-chat/api/getting-started/
markdown_url: https://smartconnections.app/smart-chat/api/getting-started.md
llms_url: https://smartconnections.app/smart-chat/api/getting-started/llms.txt
last_modified: 2026-09-04T20:55:57.370Z
usage_notes: |-
Use this page to answer questions about Getting Started.
excerpt: |-
Set up Smart Chat API Extension to answer from approved notes Use Smart Chat API Extension for answers from approved notes inside Obsidian. The model can be local or cloud. First useful result Send one bounded question and receive one completed response from the intended model. Opening Chat, selecting a model, or attaching context is preparation, not the first win. Context is request-specific…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
- title: Getting Started
url: https://smartconnections.app/connect-pro/getting-started/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
# Set up Smart Chat API Extension to answer from approved notes
Use Smart Chat API Extension for answers from approved notes inside Obsidian. The model can be local or cloud.
> [!NOTE] First useful result
> Send one bounded question and receive one completed response from the intended model. Opening Chat, selecting a model, or attaching context is preparation, not the first win.
> [!IMPORTANT] Context is request-specific
> Smart Chat does not automatically send your whole vault. Prior conversation messages can carry forward, but attached source context does not automatically copy to the next response. Review or select the sources for each request.
## Choose the correct Chat workflow
| Workflow | Use it when | Continue with |
| --- | --- | --- |
| Core provider codeblocks | You want a provider web interface such as ChatGPT or Claude inside a note. No API model is required. | [Smart Chat Getting Started](https://smartconnections.app/smart-chat/getting-started/) |
| Universal Smart Chat codeblock (Pro) | You want one `smart-chat` codeblock that can hold supported provider web-thread references or Smart Chat API Extension thread references. | [Universal Smart Chat codeblock documentation](https://smartconnections.app/docs/chat/#universal-smart-chat-codeblock-behavior-pro) |
| Smart Chat API Extension | You want a configured local or cloud model, approved sources, and saved responses inside Obsidian. | Continue below. |
## Before you start
You need:
- Smart Chat Pro installed and enabled
- one compatible chat model
- valid provider credentials or a running local model server
- one small note whose facts you can verify
- Smart Environment retrieval readiness only when using **Lookup context**
One selected model is not enough. The model must pass its available **Test** action or complete a small request. If testing returns an authorization error such as `401`, repair the credential before attaching sensitive sources.
## Get the first written answer from approved notes
### 1. Confirm one working model
1. Open **Browse Smart Plugins**.
2. Make sure that Chat Pro is enabled.
3. Open the Chat model controls in one of these locations:
- On the Chat Pro row, select **Open settings**.
- In Smart Environment settings, open the Chat model controls.
4. Under **Chat models**, select **+ New**.
5. Select an enabled provider, such as **PRO: Open Router (cloud)**, **PRO: OpenAI (cloud)**, or **PRO: Ollama (local, requires Ollama app)**.
6. Enter the required provider fields.
- For a cloud provider, enter its **API Key**.
- For Ollama, enter the **Ollama host**.
- If you use Ollama, make sure that the Ollama app is running.
7. If the provider must load its model list, select **Refresh Models**.
8. Select **Chat Model**.
9. Save or close the editor.
10. If several models exist, select **Default chat model**.
11. Make sure that the intended row is **Current**.
12. Run **Test** on the row or **Test model** in the editor.
Stop if the interface shows `MISSING MODEL`, `MISSING PROVIDER`, or a failed test. **Current** means that the model is selected. Run the model test to check credentials, provider access, or the local server.
See [Confirm or change the chat model](https://smartconnections.app/docs/chat/#confirm-or-change-the-chat-model).
### 2. Open or start a thread
Open Smart Chat in one of these ways:
- In the Command Palette under Smart Chat Pro, run **Open Smart Chat**.
- Select the Smart Chat ribbon action.
- Use an assigned hotkey.
Smart Chat resumes the active saved thread when possible. Otherwise it opens the newest non-deleted thread or creates a new one. Use **New Chat** when the resumed thread is not the right place for this test.
The toolbar also contains **Chat History**, **Chat Settings**, and **Chat Help**. Leave Chat History and Chat Manager until after the first completed response.
### 3. Write one bounded question
Open or make a small note whose claims and omissions you can verify. Write a question such as:
```prompt
Using only the attached note, answer in exactly 3 short sentences.
Sentence 1: quote the approved headline.
Sentence 2: state the promise to avoid and why.
Sentence 3: name the missing page section that should be added next.
Do not suggest changing the approved headline or invent facts.
```
Writing the question first makes context selection inspectable and gives **Lookup context** a clear target.
### 4. Attach one known source or use Lookup context
- Use **Add context** when you know the grounding note.
- Type `@` in the composer to choose context for the current response.
- Use **Lookup context** when you know the question but not the source note.
- Use a saved named context when the required sources are already collected.
On the first response, enter at least three words or more than ten characters before expecting **Lookup context** to enable. After a completed response, Lookup can be available for a continuation.
For source controls and drag paths, use [known-note context](https://smartconnections.app/docs/chat/#add-known-notes-as-context), [Lookup context](https://smartconnections.app/docs/chat/#find-context-with-lookup), and [drag and drop](https://smartconnections.app/docs/chat/#drag-notes-results-and-named-contexts-into-a-thread).
### 5. Review the source set
Keep the smallest set containing the facts and constraints needed for the question. Remove unrelated, stale, duplicated, or untrusted sources.
When context is present, the button changes from **Add context** to **Open context builder**. Select **Open context builder**. Make sure that the intended sources are present before sending.
### 6. Send only after the request is inspectable
Make sure that the intended model, question, and approved sources are visible. Select **Send** or use the default `Shift + Enter` shortcut.
If no context is attached, Smart Chat asks whether to use **Lookup context**, **Select context**, **Continue without context**, or **Cancel**. For this workflow, cancel and attach the test note. Continuing without context can be valid for general chat, but it does not test an answer from approved notes.
After sending, expect the thread status to move through `Typing` or `Generating...` before the completed response appears. An `Error` state is a failed generation, not a response.
### 7. Verify before promoting the answer
Compare the answer with the attached note. Confirm that it quotes the approved headline exactly, states the promise to avoid and why, identifies the missing section, and does not invent a fact.
The completed response can show its model, attached context, user message, and assistant answer. Treat the answer as working material until its claims match the source.
*The selected local model, one-source context boundary, question, and completed answer are visible together.*
## You know it worked when
You know the workflow is available only after both checks below succeed.
| Success layer | First-win signal |
| --- | --- |
| Product result | The intended model, approved source set, user question, and completed response are visible in one thread. |
| Useful result | The answer follows the requested three-part structure, matches the attached note, and adds no unsupported facts. |
A fluent response that contradicts the source or introduces unsupported claims is not a useful first win.
## Reopen and manage the successful thread
After the first response succeeds:
- Use **Chat History** to reopen a recent saved thread. History searches thread names and displayed recent user-prompt previews, not every word in every message.
- Press `Enter` to open the selected History item.
- Request deletion from History with the shortcut for your platform:
- On macOS, press `Cmd + Enter`.
- On other platforms, press `Ctrl + Enter`.
- Before deletion, make sure that the named thread is correct.
- Run **Open: Chat Manager view** from the Command Palette under Smart Chat Pro. Use Chat Manager for search, rename, deletion, and counts.
*Press `Enter` to open the selected saved thread.*
*Use the row actions to open, rename, or delete the intended thread. Deletion requires confirmation.*
## Recover without changing everything at once
| Problem | First recovery | Detailed guidance |
| --- | --- | --- |
| `MISSING MODEL` | Under **Chat models**, select **+ New**. Complete the provider and **Chat Model** fields. Select the model under **Default chat model**. Make sure that the row is **Current**. Run **Test** or **Test model**. | [Model setup](https://smartconnections.app/docs/chat/#confirm-or-change-the-chat-model) |
| `MISSING PROVIDER` | Repair the provider, credential, or local server configuration. Run the model test again. | [Model recovery](https://smartconnections.app/docs/chat/#recover-from-model-and-completion-errors) |
| Test returns `401` or another authorization error | Replace or repair the provider credential. Do not send note content until the test succeeds. | [Model recovery](https://smartconnections.app/docs/chat/#recover-from-model-and-completion-errors) |
| `Error` after sending | Keep the same bounded prompt. Correct the reported model or provider problem. Retry once. | [Model recovery](https://smartconnections.app/docs/chat/#recover-from-model-and-completion-errors) |
| Weak Lookup candidates | Make the question more specific. Check Environment coverage. Check the exclusions. | [Improve retrieval](https://smartconnections.app/docs/lookup/#improve-weak-or-overly-broad-results) |
| Generic or unsupported answer | Clarify the required evidence. Remove weak sources. Attach the intended source set again. Send a new request. | [Review context](https://smartconnections.app/docs/chat/#review-context-before-sending) |
| Contradiction with a source | Open the source. Compare the relevant passage. Keep the response out of trusted notes until you resolve the contradiction. | [Review responses](https://smartconnections.app/docs/chat/#review-completed-responses) |
| Expected prior sources are absent | Open **Add context** or **Open context builder**. Select the sources again. Attached context does not automatically copy to the next response. | [Review context](https://smartconnections.app/docs/chat/#review-context-before-sending) |
## Data handling and next steps
A cloud provider receives the question and source content selected for its request. A local provider can remain on the machine when that local runtime is active. Confirm the selected provider's data-handling terms before sending sensitive notes.
| Next need | Continue with |
| --- | --- |
| Full Chat controls, persistence, context, and recovery | [Smart Chat API Extension documentation](https://smartconnections.app/docs/chat/#chat-with-obsidian-notes-using-local-and-api-models) |
| Saved-thread search, rename, and deletion | [Thread management documentation](https://smartconnections.app/docs/chat/#manage-smart-chat-threads-in-obsidian) |
| Reusable reviewed source sets | [Smart Context Builder](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian) |
| Better semantic retrieval | [Smart Lookup](https://smartconnections.app/docs/lookup/#search-obsidian-notes-by-meaning-with-smart-lookup) |
| Provider-thread links stored in notes | [Smart Chat Core Getting Started](https://smartconnections.app/smart-chat/getting-started/) |
| A repeatable delegation and review loop | [Smart Loop](https://smartconnections.app/smart-loop/faq/#what-is-the-smart-loop-in-5-steps) |
| A specific vault change | [Connect Pro](https://smartconnections.app/connect-pro/getting-started/) |
---
## Faq
canonical: https://smartconnections.app/smart-chat/faq/
html_url: https://smartconnections.app/smart-chat/faq/
markdown_url: https://smartconnections.app/smart-chat/faq.md
llms_url: https://smartconnections.app/smart-chat/faq/llms.txt
last_modified: 2026-09-04T20:55:45.896Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Chat FAQs Three Chat surfaces Core provider-specific codeblocks use provider web interfaces. The universal Smart Chat codeblock can host supported provider web threads and Smart Chat API Extension thread references. Smart Chat API Extension uses a configured local or cloud model inside Obsidian. What problem does Smart Chat solve?#Smart Chat lets you use provider conversations from the…
suggested_links:
- title: Api
url: https://smartconnections.app/smart-chat/api/
- title: Codeblock
url: https://smartconnections.app/smart-chat/codeblock/
- title: Getting Started
url: https://smartconnections.app/smart-chat/getting-started/
- title: Settings
url: https://smartconnections.app/smart-chat/settings/
- title: Thread Dashboard
url: https://smartconnections.app/smart-chat/thread-dashboard/
# Smart Chat FAQs
> [!NOTE] Three Chat surfaces
> Core provider-specific codeblocks use provider web interfaces. The universal Smart Chat codeblock can host supported provider web threads and Smart Chat API Extension thread references. Smart Chat API Extension uses a configured local or cloud model inside Obsidian.
## What problem does Smart Chat solve?
[Smart Chat](https://smartconnections.app/smart-chat/) lets you use provider conversations from the notes they serve. Smart Chat API Extension lets you ask configured models inside Obsidian with context you review.
## Do I need an API key?
Core provider codeblocks do not need an API model; they use provider web interfaces and can require provider sign-in. The universal Smart Chat codeblock can host either supported provider web threads or Smart Chat API Extension threads. API Extension threads require a compatible working local model or the credentials required by their cloud provider.
## What is the difference between Core provider codeblocks, the universal Smart Chat codeblock, and Smart Chat API Extension?
| Surface | What it does | Model requirement |
| --- | --- | --- |
| Core provider codeblock | Renders one provider web interface such as ChatGPT or Claude in a note. | No API model. |
| Universal Smart Chat codeblock (Pro) | Adds one `smart-chat` codeblock that can hold supported provider web-thread references and Smart Chat API Extension thread references. | No API model for provider web threads; API Extension threads require one compatible working model. |
| Smart Chat API Extension | Runs configured local or cloud models in the Smart Chat workspace with optional approved context. | One compatible working model. |
## What counts as the first successful result?
A visible response to a concrete prompt. Installing Chat, opening a view, inserting a block, selecting a model, or saving a URL is preparation. For a note-grounded Smart Chat API Extension request, the answer must also match the approved source.
*A completed reply is the first visible result. Saving the conversation requires a recognized durable thread URL.*
## Which providers do codeblocks support?
Provider codeblocks support ChatGPT, Claude, Gemini, DeepSeek, Perplexity, Grok, Google AI Studio, Open WebUI, and Kimi. See [Supported providers and exact insert commands](https://smartconnections.app/docs/chat/#supported-providers).
## Does Smart Chat work on mobile?
Saved web thread URLs can remain useful as external bookmarks on mobile. The embedded provider surface, new-chat capture, automatic URL saving, and state actions require the desktop main Markdown pane.
## How do I save or bookmark a thread?
Add a provider codeblock to the note that owns the conversation. Use the provider until it creates a recognized durable conversation URL. Smart Chat can then write that URL into Markdown. Reopen the saved entry. Make sure that it returns to the same conversation before you rely on it.
## Why did the provider answer but the block still says new chat?
Response success and durable saving are separate. The provider returned an answer, but it has not produced a recognized durable conversation URL. Continue until a supported thread URL exists. Do not treat a temporary URL or a provider home-page URL as a saved conversation.
## What do Active and Done mean?
Active means the saved provider thread still needs attention. Done means you reviewed it and closed the loop for now. These are user-owned Markdown states.
Keep the thread Active until you can reopen it. Review the result. Change the thread to Done only after this review.
## Does Done mean the AI finished correctly?
No. Done is a user-owned tracking state. It does not validate the response.
## Is Chat Inbox the same as Chat Manager?
No. Chat Inbox reads note-attached `chat-active::` and `chat-done::` provider bookmarks. **Chat Manager** manages Smart Chat API Extension thread records. Run **Open: Chat Manager view** from the Command Palette under Smart Chat Pro.
## Does the Dataview Chat Inbox work without Pro?
Yes. Dataview can read note-attached `chat-active::` and `chat-done::` fields without Chat Pro. When Dataview is enabled, run **Insert Smart Chat thread Dataview blocks**. Otherwise, add equivalent queries manually.
## Should I delete done thread links?
No. Done thread links remain normal Markdown references until you remove them.
## Can I use local or API models?
Yes. In Chat Pro settings, use this sequence:
1. Find **Chat models**.
2. Choose **+ New**.
3. Select an enabled provider.
4. For a cloud provider, enter its **API Key**.
5. For Ollama, enter the **Ollama host**.
6. For Ollama, start the Ollama app.
7. Choose **Refresh Models** when the model list is not current.
8. Select **Chat Model**.
9. If several model rows exist, select the intended row under **Default chat model**.
10. Make sure that the row is **Current**.
11. Run **Test** or **Test model**.
**Current** means that the model is selected. It does not confirm credentials, provider access, or a running local server.
## How do I configure an API key?
Open model setup through **Open settings** on the Chat Pro Store row or through Smart Environment settings. Then use this sequence:
1. Under **Chat models**, choose **+ New**.
2. Select the cloud provider.
3. Enter the **API Key**.
4. Choose **Refresh Models** when the model list is not current.
5. Select **Chat Model**.
6. If several model rows exist, select the intended row under **Default chat model**.
7. Run **Test** or **Test model**.
Do not send note content when the test returns an authorization error such as `401`.
## Why are Send or Lookup context unavailable?
The composer is empty, no usable model is configured, or the selected model has no usable provider. For `MISSING MODEL`, follow [the Chat models setup sequence](https://smartconnections.app/docs/chat/#confirm-or-change-the-chat-model). For `MISSING PROVIDER`, repair the provider, credential, or local server. Then run **Test** again.
Enter a prompt before you expect **Send** to enable. For the first request, enter at least three words or more than ten characters. Then expect **Lookup context** to enable.
## Does Smart Chat automatically use the active note?
No. Provider codeblocks send only what you enter or upload through the provider interface. Smart Chat API Extension uses the sources selected through **Add context**, `@`, **Lookup context**, drag and drop, or Context Builder. The active note and whole vault are not attached automatically.
## Which source path should I use?
In Smart Chat API Extension, use the simplest way to add the sources you need:
- Use **Add context** or `@` when you know the source.
- Use **Lookup context** when you know the question but not the note.
- Use a saved named context or Context Builder when the required sources are already collected.
- Use **Continue without context** only when the request intentionally needs no information from your vault.
## Does attached context carry forward automatically?
No. Prior non-excluded user and assistant messages can influence a later response, but attached source context is request-specific. Use **Add context** or **Open context builder** to select the required sources again.
## Does Smart Chat store the full conversation in Markdown?
Provider codeblocks store a durable provider URL and optional Active or Done fields, not the provider transcript. Smart Chat API Extension stores its messages and attached context as Smart Plugin application data rather than ordinary Markdown note text.
## What does Include or Exclude change?
Include and Exclude change which prior messages can influence a later response in the current thread. They do not rewrite earlier messages, restore source attachments automatically, or guarantee that the next answer will use every included message.
## What leaves my device?
A provider codeblock sends what you type or upload through that provider's site. Smart Chat API Extension sends the prompt, included prior messages, and context selected for that response to the chosen model provider. A local model can keep the request on the machine when its local runtime is active. Review the selected sources and provider terms before sending sensitive material.
## Where are Smart Chat threads stored, and how should I sync them?
| Thread type | Storage and sync boundary |
| --- | --- |
| Provider-thread bookmark | Stored in Markdown and can follow the note's normal sync path. |
| Smart Chat API Extension thread | Stored as Smart Plugin application data rather than an ordinary Markdown note. |
| `.smart-env/` | Exclude it from generic third-party folder sync by default unless guidance for your specific setup says otherwise. |
Do not edit synchronized Smart Plugin application data from several devices at the same time. See the [Smart Environment FAQ](https://smartconnections.app/smart-environment/faq/#what-should-i-do-with-smart-env-in-syncthing-or-other-folder-sync-tools) for the authoritative third-party sync boundary.
## How do I keep a thread from drifting?
Review the context attached to the current response. Keep stable constraints in custom instructions. Exclude irrelevant prior exchanges from future conversation history. Reattach required sources for later requests. Do not assume that attached context carries forward.
## What is the difference between a failed generation and a bad response?
A failed generation ends in an `Error` state and produces no completed answer. Correct the model, provider, credential, server, or request problem. Retry the same bounded prompt.
A bad response completes but is generic, unsupported, or contradictory. Correct the prompt or source set, then send a new request.
## How do I reopen a recent Smart Chat API Extension thread?
Choose **Chat History** from the thread toolbar. It searches thread names and displayed recent user-prompt previews. Press `Enter` to open the selected thread. It does not search every word in every message.
## Does seeing Open, Rename, or Delete mean the action finished?
No. The control exposes the action. The resulting thread record or list state shows whether it completed.
## Can Smart Chat change my vault?
Smart Chat does not apply arbitrary AI-generated edits to your notes. It does write its own thread/bookmark state into Markdown and Smart Plugin application data. Use [Connect Pro](https://smartconnections.app/docs/connect-pro/) when you want an explicit general-purpose vault change.
## What if provider sign-in is difficult?
Complete sign-in through Obsidian Web Viewer. Return to the provider codeblock. Choose **Refresh**. Clear Web Viewer browser data only after you confirm that you can recreate the credentials and session. Otherwise, use the provider's account-recovery guidance or open a saved thread externally.
---
## Mobile
canonical: https://smartconnections.app/smart-plugins/mobile/
html_url: https://smartconnections.app/smart-plugins/mobile/
markdown_url: https://smartconnections.app/smart-plugins/mobile.md
llms_url: https://smartconnections.app/smart-plugins/mobile/llms.txt
last_modified: 2026-09-04T15:39:17.913Z
usage_notes: |-
Use this page to answer questions about Mobile.
excerpt: |-
Smart Plugins on mobile Smart Environment can defer loading on Obsidian mobile. Current detailed mobile screenshots and control placement require a fresh recapture before they should be treated as publication proof. Use the capability boundary below instead of relying on older step-by-step images. Readiness workflow Open the Smart Plugin workflow you intend to use. When Smart Environment shows…
suggested_links:
- title: Connections Early Getting Started
url: https://smartconnections.app/smart-plugins/connections-early-getting-started/
- title: Core Vs Pro
url: https://smartconnections.app/smart-plugins/core-vs-pro/
- title: Faq
url: https://smartconnections.app/smart-plugins/faq/
- title: For Entrepreneur Coders
url: https://smartconnections.app/smart-plugins/for-entrepreneur-coders/
- title: Getting Started
url: https://smartconnections.app/smart-plugins/getting-started/
# Smart Plugins on mobile
Smart Environment can defer loading on Obsidian mobile.
Current detailed mobile screenshots and control placement require a fresh recapture before they should be treated as publication proof. Use the capability boundary below instead of relying on older step-by-step images.
## Readiness workflow
1. Open the Smart Plugin workflow you intend to use.
2. When Smart Environment shows **Smart Environment not loaded** or **Idle**, select **Load Smart Environment**.
3. Keep the available loading or status view open.
4. Wait until it shows **Smart Environment ready** or **Ready**.
5. Run one small plugin action.
6. Verify the actual result separately.
A Ready status proves that the Environment reached its current ready state. It does not prove that every plugin, provider, model, source, or mobile gesture works correctly.
## Why mobile can differ from desktop
- Mobile does not provide the same persistent desktop status-bar surface.
- Startup, memory, battery, and background-execution constraints differ.
- Some desktop-only workflows require Electron webviews or direct filesystem access.
- Control placement can vary by Obsidian, operating-system, and plugin release.
Use the controls visible in the installed mobile build rather than searching for a screenshot-specific button.
## Feature boundaries to review
| Workflow | Mobile expectation |
| --- | --- |
| Connections and Lookup | Wait for Environment readiness, then verify one actual result. |
| Context using in-vault notes, blocks, and named contexts | Can remain useful; verify clipboard behavior in the destination app. |
| External files and folders | Desktop filesystem access is required. |
| Provider chat codeblocks | Saved links can remain useful, but full embedded webview behavior and automatic URL capture are desktop-dependent. |
| Pro API Chat | Treat model, storage, and sync behavior separately from Markdown provider bookmarks. |
| Connect Pro | The running Obsidian Desktop vault remains the execution side. |
## Deferred loading FAQ
### Is deferred loading on mobile a bug?
No. Smart Environment can defer loading on mobile. Select **Load Smart Environment**, wait for **Ready**, then test the first plugin result separately.
### How do I know Smart Plugins are ready?
Use the current status view until it reports **Smart Environment ready** or **Ready**. Then run the smallest useful action and inspect the result.
### Why are older mobile screenshots not shown here?
A screenshot proves only the build and state it captured. Current mobile control placement needs a fresh, matched-state recapture before it can support current instructions.
## Related pages
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
- [Smart Plugins documentation](https://smartconnections.app/docs/plugins/)
- [Smart Connections](https://smartconnections.app/docs/connections/)
- [Smart Context](https://smartconnections.app/docs/context/)
- [Smart Chat](https://smartconnections.app/docs/chat/)
---
## Faq
canonical: https://smartconnections.app/smart-templates/faq/
html_url: https://smartconnections.app/smart-templates/faq/
markdown_url: https://smartconnections.app/smart-templates/faq.md
llms_url: https://smartconnections.app/smart-templates/faq/llms.txt
last_modified: 2026-09-04T13:23:03.411Z
usage_notes: |-
Use this page to answer questions about Faq.
excerpt: |-
Smart Templates FAQs What is Smart Templates for Obsidian?#Smart Templates turns Markdown templates into reusable AI workflows. Use the current note or selected text as the starting context. Add only the context that the task needs. Select one or more templates. Add a short instruction for the specific request. Current Smart Templates 2.3 copies the completed prompt in Core and Pro. What is the…
suggested_links:
- title: Clipboard
url: https://smartconnections.app/smart-templates/clipboard/
- title: Commands
url: https://smartconnections.app/smart-templates/commands/
- title: Generate
url: https://smartconnections.app/smart-templates/generate/
- title: Getting Started
url: https://smartconnections.app/smart-templates/getting-started/
- title: Modal
url: https://smartconnections.app/smart-templates/modal/
# Smart Templates FAQs
## What is Smart Templates for Obsidian?
Smart Templates turns Markdown templates into reusable AI workflows. Use the current note or selected text as the starting context. Add only the context that the task needs. Select one or more templates. Add a short instruction for the specific request.
Current Smart Templates 2.3 copies the completed prompt in Core and Pro.
## What is the current workflow?
The shared preparation flow is:
```text
current note or selection -> context -> template -> instructions
```
Then complete the current workflow:
```text
Core and Pro -> Copy prompt -> paste -> examine -> send
```
Both editions use the same Template Context modal for preparing the request.
## Does Smart Templates replace Obsidian Templates?
No. Obsidian Templates inserts predefined content into notes. Smart Templates uses template content as reusable AI structure layered onto the current note, selection, and optional supporting context.
You can point Smart Templates at an existing Obsidian Templates folder.
## Do I need an API key?
Not for the current Smart Templates 2.3 workflow. Smart Templates builds the prompt in Obsidian and copies it so you can use ChatGPT, Claude, Gemini, Smart Chat, or another AI interface. The destination can require its own account, model, or provider configuration.
## What is the difference between Core and Pro?
Current Core and Pro surfaces use the same Template Context preparation flow and end at **Copy prompt**. This canon does not define a separate current Pro generation path.
Use the live Smart Plugins Store for current edition access. Do not infer current Generate, model-selector, or output-review controls from older Templates v1 guides.
## Where did Generate, Regenerate, Insert, and Create note go?
These controls belonged to Smart Templates v1 and are not current Smart Templates 2.3 controls. The retired workflow also included generated-output review and a Templates-specific default or per-request model selector.
Current Smart Templates ends at **Copy prompt**. The former Generate routes now lead to the [migration section](https://smartconnections.app/docs/templates/#smart-templates-generate).
## Does current Smart Templates require a configured model?
No. Smart Templates prepares and copies the request. The AI tool where you paste or send the request can require its own model, account, or provider configuration.
## How are templates detected?
Smart Templates can discover templates from configured folders, the Obsidian Templates folder, notes marked with `smart template: true`, a configured filename rule, or matching headings. Built-in templates remain available before vault-specific setup.
## Can I keep multiple templates in one note?
Yes. Matching headings can act as separate reusable templates, allowing one note to hold several related prompt structures.
## Can I use more than one template at once?
Yes. Smart Templates combines selected templates in selection order. Review the completed request when order affects section sequence or instruction priority.
## Can I use Smart Templates with ChatGPT, Claude, Gemini, or Smart Chat?
Yes. Select **Copy prompt**. Paste the request into a compatible ChatGPT, Claude, Gemini, Smart Chat, or other AI interface. Examine the pasted request before sending it.
## How is Smart Templates different from Smart Context?
[Smart Context](https://smartconnections.app/docs/context/) determines which source material belongs in the package. Smart Templates determines the reusable output structure and request instructions.
Use both when the task needs carefully assembled evidence and a repeatable result format.
## Does Smart Templates keep my notes private and local-first?
Smart Templates prepares the request in Obsidian and copies it to the clipboard. Smart Templates itself stops at the clipboard, but a configured cloud embedding or ranking provider can already receive eligible source text during Smart Environment processing. The copied prompt leaves Obsidian when you paste or send it through another application or provider-backed tool. Review the context, template, instructions, and destination before sending sensitive material.
## How do I know Copy prompt worked?
To verify **Copy prompt**:
1. Paste the copied content into a clean composer or temporary note.
2. Make sure that the intended context is present.
3. Make sure that the selected templates are present.
4. Make sure that the request-specific instructions are present.
A copy confirmation shows that the copy action is complete. The pasted content shows what Smart Templates copied.
## What if an older guide says Generate worked?
That guide describes the retired Smart Templates v1 workflow. In current Smart Templates 2.3, use **Copy prompt** and examine the complete pasted request. The destination, not Smart Templates, produces the model output.
---
## Introducing Pro Plugins
canonical: https://smartconnections.app/introducing-pro-plugins/
html_url: https://smartconnections.app/introducing-pro-plugins/
markdown_url: https://smartconnections.app/introducing-pro-plugins.md
llms_url: https://smartconnections.app/introducing-pro-plugins/llms.txt
last_modified: 2026-09-03T18:41:29.429Z
usage_notes: |-
Use this page to answer questions about Introducing Pro Plugins.
excerpt: |-
Introducing Pro plugins Pro plugins are the sustainable home for advanced Smart Plugins depth. Core Smart Plugins are built for the shortest path to trust: install, open, and get one useful result without negotiating a maze of settings. Pro plugins sit on top of that Core path. They give advanced users a focused place for deeper controls, higher-maintenance workflows, and faster-moving AI…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
# Introducing Pro plugins
*Pro plugins are the sustainable home for advanced Smart Plugins depth.*
Core Smart Plugins are built for the shortest path to trust: install, open, and get one useful result without negotiating a maze of settings.
Pro plugins sit on top of that Core path. They give advanced users a focused place for deeper controls, higher-maintenance workflows, and faster-moving AI integrations while keeping Core simple, stable, and useful for everyone.
Start with Core when you want the fastest first win. Use Pro when a real workflow asks for more depth, scale, routing, generation, or controlled action.
**Fast facts**
- Core Smart Plugins remain free and source-available.
- Pro is optional depth, not a requirement for the Core first win.
- Pro funding supports advanced maintenance while helping Core stay simpler and calmer.
- Current plan, trial, billing, and access details live on the Pro plugins page and checkout flow.
**What Pro funding unlocks next**
- Stable, repeatable install-and-go defaults for new users.
- Faster refactors to keep AI-related features aligned with provider changes.
- Deeper documentation and walkthroughs that show complete workflows instead of every edge case.
- Clearer priorities: Pro subscriptions signal which advanced workflows deserve long-term maintenance.
#### Looking ahead: I had two ways forward
To keep Smart Plugins healthy for the long term, I had to choose how to grow.
- Keep adding advanced options into the free core plugins and try to support every edge case in the same place.
- **Or** give advanced features their own home in Pro plugins, with a small investment from power users that funds deeper work on the tools.
I chose the second path. It stacks more value for people who need advanced workflows and keeps the core experience accessible for everyone else.
> The goal is simple: make the Core Smart Plugins effortless for new users, and let Pro plugins fund the time, care, and experimentation that advanced features need.
#### How Pro plugins sustain Core Smart Plugins
Pro plugins fund the advanced lane and the shared foundations that also support Core. Use the current Core and Pro product pages for today's boundary, and versioned release notes for historical changes.
- **More time for core quality**
Pro revenue offsets support costs so I can focus on polishing core features instead of constantly firefighting complex setups.
- **Stable, well documented core plugins**
With advanced options separated into Pro, the core plugins change less often, which means clearer docs, better walkthroughs, and more training videos that stay up to date.
- **Continuous maintenance for AI features**
AI providers and APIs shift quickly. Pro subscriptions fund the ongoing work to keep AI features working smoothly as the landscape changes.
- **Future-proof roadmap**
Pro funding keeps integrations current and makes it safe to test AI-related features without breaking the stable core.
- **A clear signal of what to maintain**
When people subscribe to specific Pro plugins, it shows which advanced workflows are worth long term support, testing, and refinement.
#### How to read the Core and Pro boundary
Use the current Core and Pro product pages for the capability boundary that applies today.
Use versioned release notes to understand what changed in a specific release. Do not infer that every feature has always belonged to its current edition, and do not use an announcement post as a complete migration ledger.
#### What stays free for everyone
The mission has not changed: core Smart Plugins remain free and source available.
- Core Smart Plugins are designed for a zero setup experience that just works.
- They follow the 80-20 rule: the essential 20 percent of features that give most people 80 percent of the value.
- Core features change more slowly, so your workflows break less and docs stay accurate longer.
- The Smart Environment underneath continues to power all of this.
#### What Pro plugins unlock
Pro plugins are for people who want more control over the Smart Environment and are willing to support the project in return.
Examples of what Pro plugins focus on:
- Advanced Smart Environment settings that stay aligned with upstream provider changes.
- Deeper configuration for niche or complex workflows that would confuse most new users.
- Early release and experimental features that need rapid iteration before they are ready for the core plugins.
#### For existing supporters
If you supported the project before the current Pro subscription flow, thank you. Your access is managed through the Pro access flow tied to your supporter key, account, or welcome email.
Use the current Pro plugins page, checkout/subscription page, and welcome email as the source of truth for:
- whether your access is active or grandfathered
- which Pro plugins are included while access is active
- how to manage billing, cancellation, or access issues
If anything looks inconsistent, reply to your welcome email so support can verify your access from the current account state instead of old page copy.
#### Built on the Smart Environment
Pro plugins are built on the same Smart Environment that powers the core Smart Plugins.
- The Smart Environment stays source available, local first, and focused on minimal dependencies.
- Pro revenue funds maintenance, refactoring, and new capabilities in the Smart Environment that everyone benefits from.
- Smart Plugins continue to be trusted, inspectable wrappers on top of that environment, whether you are using the free core plugins or Pro plugins.
In other words: Pro plugins do not close off the project. They are how we keep the source-available parts strong.
#### What this unlocks in the future
Introducing Pro plugins sets Smart Plugins up for the next wave of work:
- Easier onboarding for new users, with install-and-go defaults.
- Clearer roadmaps: core plugins for everyday workflows, Pro plugins for advanced systems.
- Faster development of AI related features without breaking existing setups.
- More time for documentation, walkthroughs, and examples that show complete workflows instead of explaining every possible configuration.
#### How to get Pro plugins
Start from one workflow you can evaluate in real use.
Good first Pro workflows include:
- deeper Connections control for large or noisy vaults
- Context Pro when assignments need external files, repos, PDFs, or images
- Chat Pro when you need local/API model routing and longer thread control
Use the Pro plugins page for the current plan, trial, billing, and access terms:
[Explore Pro plugins](https://smartconnections.app/pro-plugins/)
**Quick start for new users**
- Start with a Core first win.
- Pick one Pro workflow only when the Core workflow asks for more depth.
- Keep using Core if you do not need the advanced lane yet.
---
At every level, the goal is the same: keep your notes private by default, keep the tools source-available and local first, and keep improving Smart Plugins for the long term. Pro plugins are how we make that sustainable.
---
## FAQs
> [!INFO]- What are Pro plugins?
> Pro plugins are focused plugins for advanced workflows and deeper Smart Environment configuration. They sit on top of the Core Smart Plugins and give power users more control, while keeping the core plugins simple and stable for everyone.
> [!INFO]- Why introduce Pro plugins now?
> Smart Plugins grew to serve a wide range of use cases, from simple note taking to very complex AI systems. Putting all of that into the same free plugins created confusion and high support costs. Pro plugins create a dedicated home for advanced features so I can keep improving everything without burning out the project.
> [!INFO]- Will the Core Smart Plugins stay free and source-available?
> Yes. Core Smart Plugins remain free and source-available. Pro funds advanced workflows, fast-moving integrations, and deeper maintenance so Core can stay simpler and more stable.
> [!INFO]- Are you taking features away from the core plugins?
> Use the current Core and Pro product pages for today's capability boundary. Use versioned release notes for historical movement. This page does not claim that no feature has ever moved, and it does not replace a release-by-release migration record.
> [!INFO]- How does this affect existing supporters?
> Existing supporter access is managed through the current Pro access flow tied to your supporter key, account, or welcome email. Use the Pro plugins page and your welcome/subscription email as the source of truth for the access attached to your account.
> [!INFO]- Where do I find current Pro pricing, trial, and access terms?
> The current Pro plugins page and checkout flow are the source of truth for plan, trial, renewal, cancellation, and access terms. Announcement posts explain the product direction, but billing-sensitive details should match the current checkout and subscription pages.
> [!INFO]- How should I evaluate Pro?
> Start with one Pro workflow during your trial or active subscription. You do not need to explore every Pro plugin at once. Pick the workflow that already has a clear job in your vault, validate it with real work, then add more depth only when the workflow asks for it.
> [!INFO]- How do you decide what goes into core vs Pro?
> Core plugins focus on a zero setup, just works experience and the most common workflows. Pro plugins focus on deeper configuration, niche integrations, and power user features that would confuse most new users or require ongoing maintenance that needs funding.
> [!INFO]- Are Pro plugins source available?
> Pro plugins are built on top of the source-available Smart Environment. Wherever possible, shared logic lives in that environment so everyone benefits from its improvements.
> [!INFO]- Do Pro plugins send my data to your servers?
> Pro access does not automatically upload your notes. Smart Plugins are local-first by default, and provider-backed, media, generation, chat, and action workflows should be explicit choices. Review the relevant settings before sending context through an enabled integration.
> [!INFO]- What happens to experimental features?
> Experimental features may appear in Pro first when they need faster iteration or may introduce instability. Polished versions can later move into Core when they become simple, stable, and broadly useful enough for the trust-first path.
> [!INFO]- Will there still be new features in the core plugins?
> Yes. Pro plugins actually make it easier to improve the free plugins, because advanced experiments can happen in Pro first. Once a pattern is proven, the simplest version can be moved into the core plugins without adding a lot of complexity.
> [!INFO]- How will this change support and documentation?
> Support for core plugins will focus on getting new users to a "just works" setup as quickly as possible. Documentation and videos will stabilize around a smaller set of core features, while Pro documentation will be more technical and go deeper into configuration and advanced workflows.
> [!INFO]- What if I cannot afford a Pro subscription?
> Core Smart Plugins will continue to provide strong value for free. When discount, scholarship, student, or complimentary access paths are available, use the current Pro plugins page or welcome/support email as the source of truth for how to request them.
> [!INFO]- How can I share feedback or influence the roadmap?
> Supporters and Pro users can share feedback through the current support and community channels. Non-supporters can still file issues and suggestions through the open channels. Pro plugins make it easier to fund the work needed to act on that feedback.
---
## Generate
canonical: https://smartconnections.app/smart-templates/generate/
html_url: https://smartconnections.app/smart-templates/generate/
markdown_url: https://smartconnections.app/smart-templates/generate.md
llms_url: https://smartconnections.app/smart-templates/generate/llms.txt
last_modified: 2026-08-29T19:11:30.648Z
usage_notes: |-
Use this page to answer questions about Generate.
excerpt: |-
Retired Smart Templates v1 generation workflow Retired workflow Smart Templates 2.3 ends at Copy prompt. Generate, Regenerate, Insert, Create note, output review, and a per-request model selector belonged to Templates v1 and are not current Smart Templates controls. Use the current Smart Templates documentation for the supported replacement for this retired workflow. Start with Getting started…
suggested_links:
- title: Clipboard
url: https://smartconnections.app/smart-templates/clipboard/
- title: Commands
url: https://smartconnections.app/smart-templates/commands/
- title: Faq
url: https://smartconnections.app/smart-templates/faq/
- title: Getting Started
url: https://smartconnections.app/smart-templates/getting-started/
- title: Modal
url: https://smartconnections.app/smart-templates/modal/
# Retired Smart Templates v1 generation workflow
> [!WARNING] Retired workflow
> Smart Templates 2.3 ends at **Copy prompt**.
>
> **Generate**, **Regenerate**, **Insert**, **Create note**, output review, and a per-request model selector belonged to Templates v1 and are not current Smart Templates controls.
>
> Use the [current Smart Templates documentation](https://smartconnections.app/docs/templates/#smart-templates-generate) for the supported replacement for this retired workflow. Start with [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/) for the shortest verified workflow.
## Use the current Smart Templates workflow
1. Open **Smart Templates: Open template context**.
2. Confirm the source context.
3. Select one or more trusted templates.
4. Add task-specific instructions.
5. Choose **Copy prompt**.
6. Paste into a clean composer.
7. Verify the pasted request before sending it.
```text
context -> template -> instructions -> Copy prompt -> review the pasted prompt -> send in the chosen AI tool
```
Older Generate links should redirect to the retired-workflow section in the canonical documentation.
## Related documentation
- [Smart Templates](https://smartconnections.app/docs/templates/)
- [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/)
- [Smart Templates FAQ](https://smartconnections.app/smart-templates/faq/)
---
## Commands
canonical: https://smartconnections.app/smart-templates/commands/
html_url: https://smartconnections.app/smart-templates/commands/
markdown_url: https://smartconnections.app/smart-templates/commands.md
llms_url: https://smartconnections.app/smart-templates/commands/llms.txt
last_modified: 2026-08-29T19:09:31.020Z
usage_notes: |-
Use this page to answer questions about Commands.
excerpt: |-
Smart Templates commands Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Templates entry-point documentation for current behavior. Start with Getting started with Smart Templates for the shortest verified workflow. Smart Templates 2.3 keeps the command surface intentionally small. The current workflow ends at Copy prompt. The main command…
suggested_links:
- title: Clipboard
url: https://smartconnections.app/smart-templates/clipboard/
- title: Faq
url: https://smartconnections.app/smart-templates/faq/
- title: Generate
url: https://smartconnections.app/smart-templates/generate/
- title: Getting Started
url: https://smartconnections.app/smart-templates/getting-started/
- title: Modal
url: https://smartconnections.app/smart-templates/modal/
# Smart Templates commands
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Templates entry-point documentation](https://smartconnections.app/docs/templates/#smart-templates-clipboard-entry-points) for current behavior. Start with [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/) for the shortest verified workflow.
Smart Templates 2.3 keeps the command surface intentionally small. The current workflow ends at **Copy prompt**.
## The main command
- **Smart Templates: Open template context**
This opens the shared Template Context modal.
From there you can:
- start from the current note or editor selection
- add and review context
- select one or more templates
- add task-specific instructions
- copy a reviewed prompt
> [!NOTE] Current 2.3 behavior
> Smart Templates 2.3 does not expose **Generate**, **Regenerate**, **Insert**, **Create note**, an output-review modal, or a per-request Templates model selector.
>
> Those controls belonged to Templates v1. Older links should resolve to the retired-workflow section in the current documentation.
## Other entry points
### Ribbon icon
The ribbon icon opens the same shared modal.
Use it when you want a visible entry point instead of a hotkey.
### File menu
Right-click a supported Markdown or text file in the file navigator and choose:
- **Open template context**
The clicked file becomes the context seed for that invocation, even when another note is active.
## Related documentation
- [Smart Templates](https://smartconnections.app/docs/templates/)
- [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/)
- [Smart Templates settings](https://smartconnections.app/smart-templates/settings/)
---
## Just In Time Linking Obsidian Notes
canonical: https://smartconnections.app/just-in-time-linking-obsidian-notes/
html_url: https://smartconnections.app/just-in-time-linking-obsidian-notes/
markdown_url: https://smartconnections.app/just-in-time-linking-obsidian-notes.md
llms_url: https://smartconnections.app/just-in-time-linking-obsidian-notes/llms.txt
last_modified: 2026-08-29T16:47:57.866Z
usage_notes: |-
Use this page to answer questions about Just In Time Linking Obsidian Notes.
excerpt: |-
How I link notes without spending time organizing or searching in Obsidian A link belongs when it tangibly improves the note. I want useful notes to appear while I am working, mainly so I don't forget about them. I want to be able to find relevant ideas, sometimes buried in a long note, without remembering exact keywords or just-in-case organization. What I am trying to avoid I do not want…
suggested_links:
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Connect Pro
url: https://smartconnections.app/connect-pro/
- title: Core Plugins
url: https://smartconnections.app/core-plugins/
- title: Faq
url: https://smartconnections.app/faq/
# How I link notes without spending time organizing or searching in Obsidian
A link belongs when it tangibly improves the note.
I want useful notes to appear while I am working, mainly so I don't forget about them.
I want to be able to find relevant ideas, sometimes buried in a long note, without remembering exact keywords or just-in-case organization.
## What I am trying to avoid
I do not want linking to become a separate organization project.
I do not want to stop working so I can organize folders, tags, or maps of content just-in-case it helps me find something someday.
I do not want to hunt through the vault because I remember the idea of something but not the exact words.
I do not want to add links just because a result looks related.
I do not want my notes to look connected while the actual work stays unclear.
## What I do instead
I let the note I am already working in surface related notes, Smart Connections.
When I have a specific question, I use Smart Lookup.
Either way, I preview before trusting a result, then add the link only when it improves the current note.
| Need | I use | Anchor |
| ------------------------------------------------------ | ----------------- | ---------------- |
| Find notes related to the note I am already working in | Smart Connections | Current note |
| Find notes by asking a plain-language question | Smart Lookup | Question or idea |
| Find an exact word, filename, tag, or heading | Obsidian search | Exact text |
My default rule:
```md
Current note -> Smart Connections
Question -> Smart Lookup
Exact words -> Obsidian search
```
I still use search when exact text matters.
What I am avoiding is turning search, folders, tags, and manual linking into a prerequisite for progress.
## The rule I use
A result is worth linking when it helps the current note do its job.
That usually means the result helps with one of these:
- clarifying what done looks like
- remembering a prior decision
- adding useful context
Related is not enough.
The link has to improve the note.
## Why this preserves control
Smart Connections and Smart Lookup surface notes using semantic relevance.
They do not decide what belongs in my notes.
I still decide whether the result matters, and I try to write the reason in my own words.
```md
- [[useful related note]]
- Use for: [why this note matters to the current outcome]
```
If I cannot write why the link belongs, I usually do not add it. At minimum, I put them under a `Context` heading to indicate they're likely useful to AI as context.
## When I use this workflow
I use this workflow when I am trying to:
- clarify a finish line
- find prior thinking without remembering the exact words
- make sure I'm not forgetting a related idea
- build context before delegation
For Outcome notes and Smart Loop, this is especially useful when clarifying `What Done Looks Like`.
A related note can remind me of a constraint, example, source, prior decision, or edge case that might shape the outcome before I delegate the work.
## The exact workflow
> [!INFO] Tools used in this workflow
> [Smart Connections](https://smartconnections.app/smart-connections/list-feature/)
> [Smart Lookup](https://smartconnections.app/smart-lookup/search/)
### Readiness check
Before using this workflow:
- Smart Connections is installed and enabled.
- Smart Lookup is installed and enabled.
- Smart Environment has indexed enough of the vault for related results to appear.
- The current note has enough meaningful text for related results to be useful.
If the current note is empty, I add a few lines about what I'm working on for Smart Connections to become useful.
Note: Smart Lookup does not depend on the current note like Smart Connections.
### 1. Open the note
I start from the note that owns the current work, my outcome note.
For an outcome note, that might mean I am working on:
```md
## Desired outcome
## What done looks like
## Context
## Next action
```
The better the note describes the work, the more useful the related results become.
### 2. Use Smart Connections
I use Smart Connections when I want related notes to appear from the note I am already looking at.
The question is:
> What else in my vault is related to this note?
This is the workflow I use when I do not want to break focus.
I can keep writing while Smart Connections gives me peripheral vision into notes I might otherwise forget.
> [!NOTE] Watch useful notes surface while writing
> [](https://youtu.be/_i3577ti8jg?t=316)
>
> Callum starts with a thin note, adds meaningful text, and related notes appear. That is the linking workflow: let the current note surface candidates, then decide what actually belongs.
### 3. Preview before linking
The displayed score is only a relative signal. It is not a reason to blindly trust the result. So I preview promising results before I act on them.
By holding ctrl/cmd while hovering the result I can quickly see the entire note without opening another tab. This works in both Smart Connections and Smart Lookup.
### 4. Drag the useful result into the note
When a result is useful, I drag it into the note to create a link.
I try not to include the link by itself, ideally a quick line about why the linked note matters is also included.
### 5. Use Smart Lookup for search without remembering keywords
I use Smart Lookup when I remember the idea, but not the exact words, title, folder, or tag.
The question is:
> What notes in my vault match this idea?
Useful query patterns:
```txt
notes that might clarify what done looks like for this outcome
```
```txt
prior decisions about [topic]
```
```txt
examples of [constraint, pattern, problem, or source type]
```
```txt
notes related to [idea] that might be useful as context
```
Smart Lookup is especially useful when the idea may be buried inside a longer note.
### 6. Decide where the link belongs
Where the link goes depends on what the link does.
| The link helps with... | I put it under... |
| -------------------------------------------------- | ----------------------------------- |
| judging the outcome | `What Done Looks Like` |
| giving AI or future me useful source material | `Context` |
| preserving a decision, example, or constraint | `Context` or the section it changes |
| deciding what to do next | `Next action` |
| optional reading that does not change the work yet | `Reference` |
The goal is not to create a perfect link taxonomy.
The goal is to make the current note better.
## Common mistakes
### Adding links because they are related
Related is not enough.
The link should tangibly improve the current note.
### Treating Smart Connections as automatic organization
Smart Connections helps surface related notes.
Real organization happens when I decide what belongs and write why.
### Linking without previewing
Preview first. Link second.
### Building a huge related list before doing the work
A long related list can become another form of procrastination.
When I am trying to move an outcome forward, I usually want the few links that help clarify the finish line, build context, or delegate the next action.
## Related
| When I need to... | I use... |
| ----------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
| Create the note that owns the outcome | [How I use outcome notes to go from idea to outcome in Obsidian](https://smartconnections.app/getting-things-done-in-obsidian/) |
| Build a context package from useful notes | [How I build context in my Obsidian notes](https://smartconnections.app/build-context-in-obsidian-notes/) |
| Copy the note and context as an AI assignment | [How I use my notes as assignments for delegating work to AI in Obsidian](https://smartconnections.app/use-notes-with-context-to-delegate/) |
| Keep delegated work attached to the note | [How I manage chat threads in my Obsidian notes](https://smartconnections.app/manage-chats-inside-obsidian-notes/) |
| Separate trusted notes from unreviewed material | [How I maintain trust while leveraging AI in my notes](https://smartconnections.app/maintain-trust-while-using-ai-in-notes/) |
| Read the Smart Connections view docs | [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/) |
| Read the Smart Lookup docs | [Smart Lookup search workflow](https://smartconnections.app/smart-lookup/search/) |
---
## Bases
canonical: https://smartconnections.app/smart-connections/bases/
html_url: https://smartconnections.app/smart-connections/bases/
markdown_url: https://smartconnections.app/smart-connections/bases.md
llms_url: https://smartconnections.app/smart-connections/bases/llms.txt
last_modified: 2026-08-29T16:47:34.250Z
usage_notes: |-
Use this page to answer questions about Bases.
excerpt: |-
Connections in Bases Deprecated page This page is retained temporarily and may contain outdated details. Use the current Connections in Bases documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. Obsidian Bases is great at helping you manage a collection of notes. Smart Connections is great at telling you what is related to a…
suggested_links:
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
- title: Inline
url: https://smartconnections.app/smart-connections/inline/
# Connections in Bases
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Connections in Bases documentation](https://smartconnections.app/docs/connections/#smart-connections-bases) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
Obsidian Bases is great at helping you manage a collection of notes. Smart Connections is great at telling you what is related to a reference note.
The Bases integration combines both:
- Add a Connections score column to any `.base` file
- Sort and filter your collection by relevance to a reference note
- Use `score_connection` and `list_connections` directly in Bases formulas
> [!NOTE]
> Connections in Bases is part of Connections Pro. Use this page when Smart Connections already works for your vault and you have a `.base` collection you want to score or sort.
>
> If you have not seen any useful Connections result yet, start with [Getting Started](https://smartconnections.app/smart-connections/getting-started/).
> [!TLDR] The Bases + Connections loop
> 1. Open a `.base` file.
> 2. Run `Add: Connections score bases column`.
> 3. Pick a reference note or choose `Current/active file (dynamic)`.
> 4. Sort by the new score column to find relevant notes in the collection.
> 5. Optional: add a `list_connections` column to see quick link trails per row.
## Who this helps
Use this when you:
- prefer dashboards and tables for planning, triage, or review
- want to compare many notes against one reference note, such as a goal, project, outline, or research question
- want a sortable shortlist before deciding what to read, link, or review
- want to prepare a reviewed candidate set before sending notes to Smart Context
## Quick start
### Add a Connections score column
1. Open any `.base` file.
2. Open the command palette.
3. Run:
- **Smart Connections Pro: Add: Connections score bases column**
4. Select the reference note you want to compare everything against.
5. The Base refreshes and the new column appears, ready to sort.
> [!TIP]
> If you do not see the command, make sure the active note is a `.base` file.
> The command only appears when a Base is the current file.
## Pick a reference point
After running the command, you choose what the score should be relative to.
### Use a fixed note
Pick a specific note when you want a stable lens, like:
- a project hub note
- a goal or objective note
- a draft outline
- a research question note
- a decision or planning note
This is useful for planning dashboards because the meaning of the score stays consistent over time.
### Use the current note
Choose **Current/active file (dynamic)** when you want the Base to update as your active note changes.
Use this when:
- your Base is already filtered to a useful domain, such as a project folder, tag, or reading list
- you want the table to respond to the note you are currently reviewing
- you want a focused relevance dashboard rather than the full Connections list
> [!NOTE]
> Like other Bases functions that use the `this.file` property, the dynamic option works best when the Base is open in the sidebar.
## Use Connections functions in Bases formulas
You can also build your own formula columns using the two Bases functions:
- `score_connection`
- Returns a connection score.
- `list_connections`
- Returns a list of links prefixed with the connection score.
These functions appear in Bases formula autocomplete.
### Example formula usage
Use the instance form when you want the formula to read naturally for each row:
- `file.score_connection("+Projects/Project Alpha.md")`
- `file.list_connections()`
Use the global form if you prefer explicit inputs:
- `score_connection(file, "+Projects/Project Alpha.md")`
- `list_connections(file)`
### Pair score with link trails
A score column is useful for sorting, but it can be more useful when you can jump into related context immediately.
Add a `list_connections` column to show a small trail of related links beside each result.
Use this when you want to sort the collection and then open the context that explains why a row may matter.
## Understanding the score
Each score is calculated using your configured Connections scoring algorithm.
A few practical rules:
1. Treat the number as a relative signal, not an absolute grade.
2. Compare scores within the same Base and reference point.
3. Score ranges vary by embedding model, vault content, and candidate set.
4. Preview or open the row before treating it as useful.
5. If your configured algorithm uses feedback signals, hidden or pinned items can affect future ordering. Your Base reflects those changes the next time it recalculates.
> [!TIP]
> Bases is where you want the score to be actionable:
> - filter first to limit the candidate set
> - then sort by score to rank inside that set
## Workflow recipes
These workflows are for collection-level relevance: a set of rows, one or more reference notes, and a decision about what deserves attention next.
### 1) Goal-driven prioritization
Use this when you have too many notes and not enough clarity about what matters for the goal.
1. Create a Base for your project folder or tag.
2. Add a Connections score column.
3. Choose your goal note as the reference.
4. Sort descending by score.
5. Review the top rows and turn the useful ones into next actions.
Outcome: your review starts with notes most related to the goal note instead of folder order or memory.
### 2) Drafting with grounded references
Use this when you are writing and want your draft to stay anchored in existing notes.
1. Make a Base of your reference notes, sources, excerpts, or prior drafts.
2. Add a score column referenced to your outline or current draft note.
3. Sort descending.
4. Keep the Base open while you write.
5. Drag useful rows into a References section as you go.
Outcome: the draft gains a reviewed reference trail from the collection you chose.
### 3) Research triage and reading order
Use this when you have a pile of reading and need a starting sequence.
1. Make a Base for your reading list folder.
2. Add a score column referenced to your research question note.
3. Sort descending.
4. Start reviewing from the top.
5. Remove, defer, or mark rows as you decide what belongs in the current research pass.
Outcome: the reference note creates a relevance-ordered review sequence.
### 4) Compare against multiple anchors
Use this when you are balancing tradeoffs: multiple goals, multiple audiences, or multiple projects.
1. Add one score column per anchor note, such as Goal A, Goal B, Draft, or Constraints.
2. Sort by one column, then scan the others to see tradeoffs.
3. Filter to the notes that score well across the lenses you care about.
4. Review the outliers before making the decision.
Outcome: relevance becomes visible across multiple lenses instead of being hidden in a single list.
### 5) Prepare a reviewed context shortlist
Use this when you want a reviewed candidate set before packaging notes for an AI workflow.
1. Add a score column referenced to the question note, draft, or assignment note.
2. Sort descending.
3. Select the top 5 to 15 useful rows.
4. Copy links or send selected notes to Smart Context.
5. Remove noise before using the bundle.
Outcome: a reviewed candidate set that is easier to package in Smart Context or another AI workflow.
### 6) Filtered relevance dashboard for a narrow domain
Use this when the global Connections view is too broad for a recurring review.
1. Create a Base filtered to a narrow set, such as one folder, tag, project, or reading list.
2. Add a score column using `Current/active file (dynamic)`.
3. Dock the Base so it stays visible.
4. Browse notes normally.
5. Review only the rows in the chosen collection.
Outcome: a sortable relevance dashboard limited to the collection you chose.
## If Bases relevance does not help
| Symptom | Try first |
| --- | --- |
| The command does not appear | Make sure the active file is a `.base` file. |
| Scores are empty or missing across the vault | Use [Getting Started](https://smartconnections.app/smart-connections/getting-started/) to verify note eligibility and vault coverage. |
| Scores feel too broad | Filter the Base first, then sort by score inside that filtered collection. |
| A dynamic reference does not update as expected | Keep the Base open in the sidebar and confirm the reference option uses `Current/active file (dynamic)`. |
| Scores are present but ordering feels wrong repeatedly | Review [Custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/) and [Connections settings](https://smartconnections.app/smart-connections/settings/). |
| You have a question, not a table to score | Use [Smart Lookup](https://smartconnections.app/smart-lookup/search/). |
## Related pages
- Use the default list surface first: [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/)
- Ask a question instead of scoring a table: [Smart Lookup](https://smartconnections.app/smart-lookup/search/)
- Tune relevance and limits: [Connections settings](https://smartconnections.app/smart-connections/settings/)
- Tune scoring and ranking behavior: [Custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/)
- Package reviewed rows for AI workflows: [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/)
---
## 4 7
canonical: https://smartconnections.app/smart-connections/releases/4-7/
html_url: https://smartconnections.app/smart-connections/releases/4-7/
markdown_url: https://smartconnections.app/smart-connections/releases/4-7.md
llms_url: https://smartconnections.app/smart-connections/releases/4-7/llms.txt
last_modified: 2026-08-29T16:42:35.824Z
usage_notes: |-
Use this page to answer questions about 4 7.
excerpt: |-
Smart Connections Core v4.7 Related notes show up sooner. Change the anchor in place. Smart Connections still starts from the note in front of you. v4.7 removes the detour when another source should take over: choose a recent note, focus on one block, or drop a vault file onto Connections and let the list update in place. The larger upgrade is underneath the list. Smart Environment v3 gets…
suggested_links:
- title: 4 5
url: https://smartconnections.app/smart-connections/releases/4-5/
- title: 4 6
url: https://smartconnections.app/smart-connections/releases/4-6/
- title: 4 8
url: https://smartconnections.app/smart-connections/releases/4-8/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
# Smart Connections Core v4.7
## Related notes show up sooner. Change the anchor in place.
Smart Connections still starts from the note in front of you. v4.7 removes the detour when another source should take over: choose a recent note, focus on one block, or drop a vault file onto Connections and let the list update in place.
The larger upgrade is underneath the list. Smart Environment v3 gets Connections ready sooner, adds more built-in local embedding models, and lets you switch models without restarting Obsidian or deleting the embeddings you may want to return to.
> Update all installed Smart Plugins together, then restart Obsidian. Smart Connections Core v4.7 requires Smart Environment v3.
## A stronger semantic engine before the first result
Connections depends on the embedding model that turns your notes into semantic signals. v3 makes that choice less permanent and more useful:
- Start using Connections sooner after opening Obsidian.
- Choose from a broader built-in catalog, including more lightweight and multilingual local models.
- Change the active model without deleting another model's stored embeddings.
- Use the improved Environment Stats and source inspector when a note appears stale, skipped, or unexpectedly absent from results.
Learn more about the release of [Smart Environment v3](https://smartconnections.app/smart-environment/releases/3-0/).
## Change the target without changing your workspace
The active note remains the calm default. When it is not the source you want, the target menu now gives you three direct alternatives:
- Pick a recent Connections target.
- Focus on a specific block inside the current note.
- Drop another vault file onto the Connections view.
The related-note list updates around that source without making you navigate away first.
## Keep a useful result set moving
The list menu now follows the same shared action system used across Smart Environment v3. Pause or refresh discovery, open a random connection, copy the list, or continue with the reviewed sources in Context or Graph without rebuilding the set by hand.
Source menus elsewhere in the suite can also open Connections for the item you are already looking at. The workflow begins from the source, not from hunting down the right plugin command.
## Put Connections where it helps
Footer Connections now uses a configurable display component instead of a single graph on/off switch. Choose the supported presentation that fits the note surface and screen size. Scores also follow the final displayed ranking score when another ranking step provides one, so the number beside a result better matches the order you see.
## Before / After
| Before | With Smart Connections Core v4.7 |
| --- | --- |
| Connections could take longer to become ready after startup. | Smart Environment v3 reduces blocking and repeated startup work. |
| Trying another embedding model felt like committing to a rebuild. | Switch models without deleting the previous model's stored embeddings. |
| Retargeting often meant navigating to another note first. | Choose a recent note, a current-note block, or drop a file onto the view. |
| A useful result list could become a dead end. | Continue the reviewed sources through consistent Context and Graph actions. |
| Footer display was controlled by one graph toggle. | Choose the supported Connections component that fits the surface. |
## Supporting improvements
- More consistent list, result, command, ribbon, pause, and random-connection actions.
- A shared source action for opening Connections from supported Smart Plugin menus.
- Better score display after optional ranking.
- Improved vector compatibility and performance reporting through Smart Environment v3.
## Learn more
- [Smart Connections overview](https://smartconnections.app/smart-connections/)
- [Smart Connections documentation](https://smartconnections.app/docs/connections/)
- [Smart Connections getting started](https://smartconnections.app/smart-connections/getting-started/)
- [Smart Connections FAQ](https://smartconnections.app/smart-connections/faq/)
## Release notes
### `v4.7.2`
Updated: Smart Environment
Updated: minimum Obsidian app version to 1.8.7
Added: Connections list display configuration
### `v4.7.0`
improved: score handling uses score_display if available
improved: allow configurable component for connections footer
improved: enhance score calculation logic and update is_vec function to support ArrayBuffer views
improved: Connections list menu now handled using Smart Environment menu actions pattern to enable deeper integration and extendability
improved: Connections list item menu now handled using Smart Environment menu actions pattern to enable deeper integration and extendability
Added: control connections anchor/target note from menu actions. Select new target from recent connections history and blocks inside the current note.
Improved: random and pause connections features migrated to actions pattern.
Added: view connections action for source menus.
migrated: ribbon icons to actions architecture
Migrated commands to actions command architecture
Added: drag-and-drop functionality for connections to update connections target from dropped file
Enhance performance logging in get_results method and emit event with elapsed time
Add footer connections list component configuration and remove show_graph setting in favor of component selection in settings
Updated: Smart Environment v3
### v4.7.0
- Added target selection from recent Connections history and blocks in the current note.
- Added drag-and-drop retargeting from vault files.
- Added a shared source action for opening Connections for supported items.
- Unified list, result, command, ribbon, pause, and random-connection actions on the Smart Environment v3 action system.
- Added configurable Footer Connections components and replaced the previous graph toggle with component selection.
- Improved the score shown after optional result ranking.
- Improved Smart Environment v3 compatibility and performance reporting.
- Updated Smart Connections Core to Smart Environment Core v3.1.0.
- Requires Smart Environment v3.0.0 or newer.
---
## 4 5
canonical: https://smartconnections.app/smart-connections/releases/4-5/
html_url: https://smartconnections.app/smart-connections/releases/4-5/
markdown_url: https://smartconnections.app/smart-connections/releases/4-5.md
llms_url: https://smartconnections.app/smart-connections/releases/4-5/llms.txt
last_modified: 2026-08-29T16:39:48.624Z
usage_notes: |-
Use this page to answer questions about 4 5.
excerpt: |-
Smart Connections v4.5 Connections Footer is now a Core feature! Footer connections are now included in Smart Connections Core, bringing the most mobile-friendly and no-panel writing surface to every install. Connections Pro continues to add inline discovery, Bases workflows, and advanced ranking control. Place your connections list in the footer of every note. Toggle footer connections from the…
suggested_links:
- title: 4 6
url: https://smartconnections.app/smart-connections/releases/4-6/
- title: 4 7
url: https://smartconnections.app/smart-connections/releases/4-7/
- title: 4 8
url: https://smartconnections.app/smart-connections/releases/4-8/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
# Smart Connections `v4.5`
## Connections Footer is now a Core feature!
[Footer connections](https://smartconnections.app/smart-connections/footer/) are now included in Smart Connections Core, bringing the most mobile-friendly and no-panel writing surface to every install. Connections Pro continues to add inline discovery, Bases workflows, and advanced ranking control.
- Place your connections list in the footer of every note.
- Toggle footer connections from the command palette (hotkey), ribbon icon, or settings.
## Recent highlights
- Connections lists can now open with a graph view.
- [Substrate Update.](https://smartconnections.app/smart-plugins/substrate-update/)
## Release notes
### `v4.5.3`
refactor: update D3 load to use ESM bundle and improve loading logic
refactor: update CSS selectors for improved specificity and remove redundant styles
### `v4.5.2`
- Updated Smart Environment to `v2.4.6` for handling Obsidian API changes
### `v4.5.1`
- Smart Lookup has been removed from Smart Connections and is now available as a standalone plugin in the Obsidian Community Plugin Store.
- remove unused restart_plugin method
- remove periodic update check
### `v4.5.0`
- Fixed: transformers embedding model should fallback to non-GPU v4 usage and subsequently v3 if that still fails
- Fixed: should only calculate connections results once (improves performance)
---
## Codeblock
canonical: https://smartconnections.app/smart-chat/codeblock/
html_url: https://smartconnections.app/smart-chat/codeblock/
markdown_url: https://smartconnections.app/smart-chat/codeblock.md
llms_url: https://smartconnections.app/smart-chat/codeblock/llms.txt
last_modified: 2026-08-29T16:23:43.443Z
usage_notes: |-
Use this page to answer questions about Codeblock.
excerpt: |-
Smart Chat codeblocks Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Chat codeblock documentation for current behavior. Start with Getting started with Smart Chat for the shortest verified workflow. Smart Chat Notes lets you embed real, web-based chat UIs (ChatGPT, Claude, Gemini, Grok, Perplexity, DeepSeek, AI Studio, Open WebUI, Kimi)…
suggested_links:
- title: Api
url: https://smartconnections.app/smart-chat/api/
- title: Faq
url: https://smartconnections.app/smart-chat/faq/
- title: Getting Started
url: https://smartconnections.app/smart-chat/getting-started/
- title: Settings
url: https://smartconnections.app/smart-chat/settings/
- title: Thread Dashboard
url: https://smartconnections.app/smart-chat/thread-dashboard/
# Smart Chat codeblocks
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Chat codeblock documentation](https://smartconnections.app/docs/chat/#smart-chat-codeblocks) for current behavior. Start with [Getting started with Smart Chat](https://smartconnections.app/smart-chat/getting-started/) for the shortest verified workflow.
Smart Chat Notes lets you embed real, web-based chat UIs (ChatGPT, Claude, Gemini, Grok, Perplexity, DeepSeek, AI Studio, Open WebUI, Kimi) directly inside the note they belong to.
The core shift:
your chat threads stop being "lost in browser tabs" and become trackable artifacts inside your vault.
You can bookmark threads per note, return later, and mark them done when ready.
- Threads are saved back into the codeblock as normal markdown lines
- Threads can be marked Active or Done (with timestamps)
- Dataview can turn those lines into dashboards
If you want to send large, curated vault context to a thread from the exact note it belongs to, pair this with Smart Context Builder (or use the built-in Build context button):
[Smart Context Builder](https://smartconnections.app/smart-context/builder/)
## What this solves
If any of these are true, this feature will feel immediately useful:
- "I have great AI outputs, but I cannot find the thread later."
- "My project knowledge is in Obsidian, but my AI conversations live elsewhere."
- "I open 10 tabs per project and lose track of what I was waiting on."
- "I ask something that takes time, then forget to come back for the result."
- "I want to keep AI work next to the work it references."
This is not about replacing your chat provider.
It is about organizing and operationalizing your threads inside your notes.
## Mental model: the note is the work item
The note is the durable work item.
The thread is the saved AI work attached to that note.
- If you are waiting on the thread: keep it bookmarked in the relevant note.
- When the thread is resolved: click Mark done (or stop tracking it).
- The note becomes a lightweight "async inbox" for the project without making the thread the final deliverable.
This is how you get the asynchronous upside without losing control.
### Active vs Done
- Active means: still in progress, still relevant, still waiting on something.
- Done means: the conversation is done (and you can always re-activate it later).
## Two-minute quickstart
1. Open the note where the work actually lives (project hub, decision note, draft, bug report, meeting note).
2. Insert a Smart Chat codeblock for your provider (via Command Palette).
3. Start a new chat or open an existing thread inside the embed.
4. Once the provider URL becomes a real thread link, Smart Chat auto-saves it into the codeblock (pinning it to this note).
5. Optional (high leverage): click Build context to curate vault context for this thread before you send the next prompt.
6. When you are done with the thread, click Mark done.
> [!Tip]
> You do not need a "perfect system" first.
> Start by embedding one thread in one important note.
> The habit is the product: always pin the thread to the note it is about.
> Start by embedding one thread in one important note.
> [!NOTE] Watch a provider thread become note-attached work
> [](https://youtu.be/_i3577ti8jg?t=1118)
>
> Callum inserts a provider chat codeblock, sends selected context, and keeps the thread state inside the project note where the work lives.
## What the embedded controls mean
Across providers, the embedded UI adds a small control layer so the note stays in charge.
Top controls:
- Thread dropdown (Thread URL list): saved/bookmarked threads for this note
- State chip: Unsaved, Active, Done
- Mark done / Mark active: toggle state for the current thread
- Help icon: opens this documentation page
Bottom controls:
- Refresh: reload the embed if the provider UI gets stuck
- Build context: open Smart Context Builder scoped to this thread
- Open in browser: jump out to a full tab when you need it
- Copy link: copy the current thread URL
- Grow / Contain: expand the embed for comfortable reading/writing
Your markdown remains the source of truth: the note stores thread links (and status), not the provider UI.
That means threads are searchable, linkable, and dashboardable like everything else in Obsidian.
Example of what gets saved into the codeblock:
````md
```smart-chat
chat-active:: 1767302492 https://chatgpt.com/c/6956e559-8060-8329-8150-7167e477c05a
chat-done:: 1767132305 https://chatgpt.com/c/69544c91-0c78-832e-8e49-d21049a33e51
```
````
## ChatGPT example: new -> active -> done
New chat (not yet saved):
Active thread (saved and tracked):
Done thread (completed conversation, can be re-activated):
### Active vs Done
- Active means the thread is still in progress, still relevant, or still waiting for review.
- Done means the thread is resolved for now.
## Supported providers
Use the codeblock that matches your provider:
| Codeblock | Provider |
| --- | --- |
| `smart-chatgpt` | ChatGPT (also recognizes Codex and Sora links) |
| `smart-claude` | Claude |
| `smart-gemini` | Gemini |
| `smart-grok` | Grok |
| `smart-perplexity` | Perplexity |
| `smart-deepseek` | DeepSeek |
| `smart-aistudio` | Google AI Studio |
| `smart-openwebui` | Open WebUI |
| `smart-kimi` | Kimi |
## Supported platforms (annotated examples)
Below are examples of the same workflow across different providers.
- Green: saved thread URLs inside the codeblock
- Magenta: Mark done
- Orange: size/visibility controls (Grow)
- Cyan: utility controls (Refresh / Open / Copy)
Provider UIs change frequently, but the workflow stays the same: save the thread link into the note, then track it until you are done.
### ChatGPT
### Claude
### Grok
### Gemini
### Perplexity
### Deepseek
### AI Studio
## Dashboards with Dataview
Because thread state is stored as `chat-active::` and `chat-done::` in your notes, Dataview can turn your vault into an async dashboard.
Dataview plugin:
[Install Dataview](https://obsidian.md/plugins?id=dataview)
In Progress:
````md
```dataview
LIST WITHOUT ID file.link
WHERE chat-active
SORT file.mtime DESC
```
````
Completed:
````md
```dataview
LIST WITHOUT ID file.link
WHERE chat-done
SORT file.mtime DESC
```
````
### Show totals
If you want a quick "how much am I actually using this" metric across your vault, use this DataviewJS snippet:
````md
```dataviewjs
function count_field_value_instances(value) {
if (value === null || value === undefined) return 0;
return Array.isArray(value) ? value.length : 1;
}
const pages = dv.pages()
.where(p => p["chat-active"] || p["chat-done"])
.array();
const totals = pages.reduce((acc, page) => {
acc.active_total += count_field_value_instances(page["chat-active"]);
acc.done_total += count_field_value_instances(page["chat-done"]);
return acc;
}, { active_total: 0, done_total: 0 });
dv.paragraph(`**Total chat-active instances:** ${totals.active_total}`);
dv.paragraph(`**Total chat-done instances:** ${totals.done_total}`);
dv.paragraph(`**Total tracked threads:** ${totals.active_total + totals.done_total}`);
```
````
Tip: To scope this to a folder, replace `dv.pages()` with `dv.pages('"YourFolderName"')`.
## Troubleshooting
### Having trouble with signing in (common gotchas)?
Some providers use sign-in flows that are easier to complete in Obsidian's Web viewer first:
* Claude sign-in: enable Obsidian's Web viewer core plugin, log in there, then return and Refresh the embed.
* Google sign-in (Gemini / AI Studio): same idea - complete sign-in in Web viewer, then Refresh.
* AI Studio: after sending the first message, click Save in AI Studio so Smart Chat can capture the thread URL.
#### Still having trouble?
Is the "prove you are a human" continuously displayed?
Try clearing the Web Viewer browser data in the settings.
### Does it work on mobile?
These embeds rely on Obsidian desktop (Electron) webview support.
On mobile (iOS/Android), the embedded web UI cannot render the same way.
Practical takeaway:
* Desktop: full embedded chat UI + controls
* Mobile: treat the codeblock as a thread bookmark list (still useful for continuity)
Technical background:
[https://github.com/brianpetro/smart-chatgpt-obsidian/issues/7#issuecomment-3505152432](https://github.com/brianpetro/smart-chatgpt-obsidian/issues/7#issuecomment-3505152432)
## Next steps
* Add one embedded chat codeblock to your highest-value project hub note.
* Create a "Chat Inbox" dashboard note with Dataview.
* If you routinely reuse context, build your first named context pack:
[Smart Context Builder](https://smartconnections.app/smart-context/builder/)
If you want the broader Smart Chat story (and how this relates to Pro capabilities), start here:
[https://smartconnections.app/smart-chat/](https://smartconnections.app/smart-chat/)
If you are deciding between Core and Pro:
[https://smartconnections.app/pro-plugins/](https://smartconnections.app/pro-plugins/)
## Related pages
- [Smart Chat overview](https://smartconnections.app/smart-chat/)
- [Get started with Smart Chat](https://smartconnections.app/smart-chat/getting-started/)
- [Use local and API models in Smart Chat](https://smartconnections.app/smart-chat/api/)
- [Configure Smart Chat settings and defaults](https://smartconnections.app/smart-chat/settings/)
- [Manage Smart Chat threads from one dashboard](https://smartconnections.app/smart-chat/thread-manager/)
- [Build a Chat Inbox with Dataview](https://smartconnections.app/smart-chat/thread-dashboard/)
- [Build and save reusable context packs](https://smartconnections.app/smart-context/builder/)
- [Attach a Smart Context codeblock to a note](https://smartconnections.app/smart-context/codeblock/)
- [Explore the Smart Connections view](https://smartconnections.app/smart-connections/list-feature/)
---
## Search
canonical: https://smartconnections.app/smart-lookup/search/
html_url: https://smartconnections.app/smart-lookup/search/
markdown_url: https://smartconnections.app/smart-lookup/search.md
llms_url: https://smartconnections.app/smart-lookup/search/llms.txt
last_modified: 2026-08-26T17:22:43.189Z
usage_notes: |-
Use this page to answer questions about Search.
excerpt: |-
Find notes by meaning with Smart Lookup Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Lookup search documentation for current behavior. Start with Getting started with Smart Lookup for the shortest verified workflow. Smart Lookup searches your Obsidian vault by meaning, not only by exact words. Type an idea, topic, or question. Lookup…
suggested_links:
- title: Faq
url: https://smartconnections.app/smart-lookup/faq/
- title: Getting Started
url: https://smartconnections.app/smart-lookup/getting-started/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
# Find notes by meaning with Smart Lookup
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Lookup search documentation](https://smartconnections.app/docs/lookup/#smart-lookup-search) for current behavior. Start with [Getting started with Smart Lookup](https://smartconnections.app/smart-lookup/getting-started/) for the shortest verified workflow.
Smart Lookup searches your Obsidian vault by meaning, not only by exact words.
Type an idea, topic, or question. Lookup returns a ranked list of notes or note sections that may contain useful source material, even when your notes use different wording.
> [!IMPORTANT] Smart Lookup finds sources. It does not write an answer.
> To receive a written answer based on your notes, first verify the useful results, then continue to [Get a written answer from the notes you found](#context-handoff).
> [!TLDR] First useful result
> 1. Click the Smart Lookup magnifying-glass icon, or run `Smart Lookup: Open: Lookup view`.
> 2. Describe the idea you want to find.
> 3. Leave **Auto-submit** enabled, or click **Lookup**.
> 4. Expand or preview 1 to 3 promising results.
> 5. Open the result that helps your work.
You know it worked when:
> One result contains useful material related to your query, even though the note may use different words.
## Quick start
### 1. Open Smart Lookup
Use either route:
- Click the Smart Lookup magnifying-glass icon in Obsidian's ribbon.
- Press `Ctrl/Cmd + P`, then run:
```obsidian-command
Smart Lookup: Open: Lookup view
```
The Smart Lookup view opens in the sidebar with a query box that says:
```txt
Describe the idea, topic, or question you want to explore...
```
> [!NOTE] If the exact command is missing
> Confirm Smart Connections is enabled, then restart Obsidian after installing or updating it.
>
> If the command still does not appear, the instructions do not match your installed build. Do not keep looking for a control that is not there. Use Help opened from that build or verify which version supports Smart Lookup.
### 2. Describe what you want to find
Use ordinary language. A useful pattern is:
```txt
subject + context + distinguishing detail
```
Examples:
```txt
preventing information overload while researching
```
```txt
prior decisions about the Smart Context Builder
```
```txt
customer onboarding friction and activation
```
A question also works:
```txt
What have I written about information overload and focus?
```
Lookup uses the question to find likely source notes. The result is still a list, not a composed answer.
### 3. Run the search
- With **Auto-submit** enabled, pause briefly after typing and the results update automatically.
- With **Auto-submit** disabled, click **Lookup**.
Use one clear intent per query. A longer prompt is not automatically a better search.
### 4. Verify the strongest candidates
Start with the first 1 to 3 results.
1. Click the arrow beside a promising result to expand its excerpt.
2. Read enough to confirm why it matched.
3. On desktop, hold `Ctrl/Cmd` while hovering the result to use Obsidian's preview when available.
4. Click the result title when you need the full note.
Treat each result as a lead to inspect, not a conclusion to trust.
### 5. Use the result
A useful result should change the work, not remain in a result list.
You can:
- recover an idea, decision, example, or constraint from the source
- add a normal Obsidian link to the note you are working in
- drag one verified result into a supported destination where available
- open the list-level **More actions** menu and choose **Open in Context Builder** to review the current result set
- choose **Explore in Smart Graph** when the result set should become a graph scope
A purposeful link is easier to reuse:
```md
## Context
- [[Useful result note]]
- Use for: [why this note matters to the current question or outcome]
```
> [!NOTE] Lookup-to-Context behavior
> There is no verified per-result button literally named **Add to Context**.
>
> The current list-level handoff is **More actions** -> **Open in Context Builder**. Use that action to review the result set before copying or sending it.
## Choose the result you want
| What you want | Use | What it returns |
| --- | --- | --- |
| Notes that match an idea you type | **Smart Lookup** | Ranked notes or note sections |
| Notes related to the note already open | [Smart Connections](https://smartconnections.app/smart-connections/list-feature/) | Ranked results based on the current note |
| An exact word, title, heading, tag, syntax, or regex | Obsidian search | Exact text matches |
| A written answer based on selected notes | [Smart Chat API Extension](https://smartconnections.app/smart-chat/api/getting-started/) or [Smart Context](https://smartconnections.app/smart-context/getting-started/) plus your AI tool | A generated answer after source review |
```txt
Idea or question -> Smart Lookup finds sources.
Current note -> Smart Connections finds related sources.
Exact words -> Obsidian search finds exact matches.
Written answer -> AI uses the sources you approve.
```
## Write better Lookup queries
Describe the material you want to find, not the report you want an AI to write afterward.
| Too broad | More useful |
| --- | --- |
| `focus` | `focus while researching information overload` |
| `GTD` | `getting things done task prioritization` |
| `Smart Context` | `prior decisions about the Smart Context Builder` |
| `onboarding` | `customer onboarding friction and activation` |
Avoid using Lookup as though it were a chat prompt:
```txt
Summarize my notes and recommend the best next actions.
```
Search for the source material instead:
```txt
notes about project risks, unresolved decisions, and next actions
```
Then give the summarizing or recommendation instruction to the AI workflow that receives the verified notes.
If results are too broad, add one distinguishing detail. If they are too narrow, remove the least important detail.
## How to read Lookup results
Depending on your result settings, a row may show:
| Detail | Meaning |
| --- | --- |
| Score | Relative similarity to the current query |
| Note title | The source file containing the result |
| Heading path | The section where the matched material appears |
| Line range | The approximate source lines for a section result |
| Expand arrow | Opens an inline excerpt for verification |
Use the score to decide what to inspect first.
Do not assume that a high score makes the source correct, that the first result is the only useful result, or that scores from different queries are directly comparable.
Preview before opening. Open before trusting. Link only when the result improves the current work.
## Get a written answer from the notes you found
For a vault with hundreds or thousands of notes, use this flow:
```txt
search the local index -> verify a small source set -> send only approved sources -> review the answer
```
Do not send all 1,500 notes to a model for every question.
### Answer inside Obsidian with Smart Chat API Extension (Pro)
1. Open [Smart Chat API Extension](https://smartconnections.app/smart-chat/api/getting-started/).
2. Add the notes you verified, or use Smart Chat's **Lookup context** action.
3. Review the proposed source list and remove anything that should not guide the answer.
4. Ask the question.
5. Compare the response with the attached sources before adding anything to a trusted note.
### Answer with the AI tool you already use
1. Verify the useful Lookup results.
2. Open the list-level **More actions** menu.
3. Choose **Open in Context Builder**.
4. Remove anything unrelated, stale, duplicated, or untrusted.
5. Copy the reviewed context bundle.
6. Paste it into ChatGPT, Claude, Gemini, or another AI tool and ask the question.
Prompt starter:
```prompt
Use only the attached notes as source material.
Question: [your question]
Name the notes or sections you relied on.
If the sources do not contain enough information, identify the exact missing note, fact, or decision instead of guessing.
```
> [!NOTE] Data boundary
> Core Smart Lookup can search the local Smart Environment index after your notes are prepared.
>
> A cloud provider receives note content only when you explicitly attach, copy, paste, or send approved sources through that provider-backed workflow.
## Useful workflows
### Recover an idea without remembering its words
1. Describe the idea.
2. Expand the top results.
3. Open the note that confirms the match.
4. Add the recovered insight or a purposeful link to the current work.
### Find prior decisions before repeating work
1. Search for the topic plus `decision`, `tradeoff`, or the project name.
2. Preview the strongest candidates.
3. Open the source containing the actual reasoning.
4. Add the useful decision or constraint to the current note in your own words.
### Prepare source-grounded AI context
1. Search for the evidence the assignment needs.
2. Verify the useful results.
3. Use **More actions** -> **Open in Context Builder**.
4. Remove noise before copying.
5. Ask for the outcome only after the source set is reviewable.
## When results are weak
| What happened | Try first | Why |
| --- | --- | --- |
| Results are too broad | Add 1 or 2 distinguishing details. | The query needs more context to separate nearby topics. |
| Results are too narrow | Remove one detail or use a broader phrase. | Too many constraints can hide useful neighboring notes. |
| You need an exact phrase | Use Obsidian search. | Exact words, titles, tags, headings, and regex are lexical jobs. |
| You are starting from one open note | Use Smart Connections. | The current note should determine the results. |
| Results run while you are typing | Turn off **Auto-submit** and click **Lookup** manually. | Auto-submit runs after a short pause. |
| Results do not update after typing | Enable **Auto-submit** or click **Lookup**. | The query has not been submitted yet. |
| Whole-note results hide the useful passage | Change the result type from Sources to Blocks. | Blocks return smaller note sections. |
| Results are empty or stale across several queries | Check Smart Environment readiness, exclusions, and embedding progress. | Lookup depends on prepared source embeddings. |
If this is your first run, verify readiness before judging result quality:
- [Smart Milestones](https://smartconnections.app/smart-environment/milestones/)
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
## Optional: Sources vs Blocks
Start with the default result type and find one useful result before tuning.
Lookup reuses result-list configuration from Smart Connections, so some changes can affect both retrieval surfaces.
| Result type | Best when | Tradeoff |
| --- | --- | --- |
| Sources | You want broad whole-note discovery. | Easier to scan, but you may need to open the note to find the exact passage. |
| Blocks | Long notes hide the useful section. | More precise, but creates more granular candidates and depends on block preparation. |
Start with **Sources**. Use **Blocks** when the correct note appears but the useful section is buried inside it.
For the current controls, see [Smart Connections settings](https://smartconnections.app/smart-connections/settings/#lookup-lists).
## FAQ
### Why did a result appear without my exact words?
That is the purpose of searching by meaning. Use Obsidian search when exact wording matters.
### Should I trust the first result?
No. Expand or preview it, open the source when needed, and decide whether it actually helps the work.
### Where is Add to Context?
There is no verified per-result button literally named **Add to Context**. Open the list-level **More actions** menu and choose **Open in Context Builder** to review the current result set. Drag one verified result into another supported surface only when you want that single result.
## Related pages
- [Find notes related to the current note](https://smartconnections.app/smart-connections/list-feature/)
- [Build a reviewed source set with Smart Context](https://smartconnections.app/smart-context/builder/)
- [Get a written answer with Smart Chat API Extension](https://smartconnections.app/smart-chat/api/getting-started/)
- [Check Smart Environment readiness](https://smartconnections.app/smart-environment/settings/)
---
## Codeblock
canonical: https://smartconnections.app/smart-context/codeblock/
html_url: https://smartconnections.app/smart-context/codeblock/
markdown_url: https://smartconnections.app/smart-context/codeblock.md
llms_url: https://smartconnections.app/smart-context/codeblock/llms.txt
last_modified: 2026-08-26T17:19:56.108Z
usage_notes: |-
Use this page to answer questions about Codeblock.
excerpt: |-
Smart Context codeblock Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Context codeblock documentation for current behavior. Start with Getting started with Smart Context for the shortest verified workflow. The Smart Context Codeblock keeps a visible context manifest inside the note that needs it. Use it when the note body contains the…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-context/bases/
- title: Builder
url: https://smartconnections.app/smart-context/builder/
- title: Canvas
url: https://smartconnections.app/smart-context/canvas/
- title: Clipboard
url: https://smartconnections.app/smart-context/clipboard/
- title: Faq
url: https://smartconnections.app/smart-context/faq/
# Smart Context codeblock
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Context codeblock documentation](https://smartconnections.app/docs/context/#smart-context-codeblock) for current behavior. Start with [Getting started with Smart Context](https://smartconnections.app/smart-context/getting-started/) for the shortest verified workflow.
The Smart Context Codeblock keeps a visible context manifest inside the note that needs it.
Use it when the note body contains the instructions, and the codeblock contains the exact notes or named contexts that should travel with those instructions.
## Insert the codeblock
Core command:
- `Smart Context: Insert codeblock (add notes & named contexts)`
Pro command:
- `Smart Context: Insert codeblock (add external files & named contexts)`
Run the command from any note to insert the block and start editing it in place.
## Supported aliases
All three aliases render the same native UI:
````md
```ctx
Projects/Alpha/Brief.md
ctx:: Weekly planning
```
````
- `ctx`
- `context`
- `smart-context`
The original alias is preserved when the block is rewritten.
## Empty state and actions
An empty codeblock renders a compact Smart Context UI directly in the note.
The codeblock stays in the note, so you can manage local context without switching to another surface.
## Open the codeblock menu
The menu exposes the main actions for the current codeblock.
The current menu shown here includes:
- `Create named context`
- `Open context builder`
- `Copy context to clipboard`
- `Open named contexts dashboard`
- `Help`
`Create named context` is the fastest way to turn a local codeblock working set into one reusable named context.
## When the block has context items
Once the block contains items, the UI shows live size estimates in the action bar.
The copy button becomes the fastest way to export the current codeblock context.
After copy, the notice confirms the export immediately inside Obsidian.
## What gets written into the block
Whole sources are stored as one line per source:
A named context is stored as a single named-context line:
That keeps the raw markdown readable and easy to diff.
> [!NOTE] Watch a context manifest live inside the note
> [](https://youtu.be/_i3577ti8jg?t=656)
>
> Callum inserts a Smart Context codeblock so the project note can carry the exact context list instead of depending on memory or a hidden working set.
Example:
````md
```ctx
Personal Productivity/Tools & Systems/Productivity Apps & Tools.md
PKM/Advanced/Integrating Tools and Techniques.md
PKM/Basics/Capture Tools.md
```
````
Named-context example:
````md
```ctx
ctx:: Smart Context 2026-03-26
```
````
## Copy behavior
Use the codeblock copy action when you want only the codeblock context.
Use the regular current-note copy command when you want:
* the current note
* the hydrated codeblock context
* one combined clipboard bundle
In that current-note flow, codeblock items are treated as depth `0`.
## Pro adds external files and folders
On Pro, the same codeblock surface also accepts:
* external files
* external folders
* exclusion lines starting with `!`
That lets one note carry both the writing instructions and the external repo files that should travel with them.
## Related pages
- [Build and save reusable context packs](https://smartconnections.app/smart-context/builder/)
- [Reopen and reuse saved named contexts](https://smartconnections.app/smart-context/builder/named/)
- [Copy notes and folders as AI-ready context](https://smartconnections.app/smart-context/clipboard/)
- [Copy the current note with link-depth control](https://smartconnections.app/smart-context/clipboard/current/)
- [Use file navigator actions to copy notes and folders as context](https://smartconnections.app/smart-context/file-nav-actions/)
- [Copy context from Obsidian Canvas files](https://smartconnections.app/smart-context/canvas/)
- [Copy context with images and PDFs](https://smartconnections.app/smart-context/clipboard/media/)
- [Control Smart Context templates and export format](https://smartconnections.app/smart-context/settings/)
---
## Settings
canonical: https://smartconnections.app/smart-connections/settings/
html_url: https://smartconnections.app/smart-connections/settings/
markdown_url: https://smartconnections.app/smart-connections/settings.md
llms_url: https://smartconnections.app/smart-connections/settings/llms.txt
last_modified: 2026-08-26T15:55:09.107Z
usage_notes: |-
Use this page to answer questions about Settings.
excerpt: |-
Smart Connections settings Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Connections documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. Do the first win before tuning Open one real note, open Connections, preview one result, then drag or open it if useful. Tune…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
# Smart Connections settings
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Connections documentation](https://smartconnections.app/docs/connections/) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
> [!TIP] Do the first win before tuning
> Open one real note, open Connections, preview one result, then drag or open it if useful.
>
> Tune settings after you have one real result to improve.
Tune Smart Connections for speed (limits and filters), precision (sources vs blocks, scoring/reranking), and UI defaults (how results render in the main list, inline, and footer). These settings apply to both Connections (note/block suggestions) and Lookup (semantic search).
- **Connections**: suggestions related to the current note (or block), shown as a list and, where enabled by the selected component, visual surfaces.
- **Lookup**: an explicit search/lookup view that returns similar sources or blocks.
| Setting | What it changes | When to use it |
| --- | --- | --- |
| [Connection results type](https://smartconnections.app/smart-connections/settings/#connections-lists) | Sources = whole notes; Blocks = sections/blocks (more granular) | Choose Sources for speed and overview; choose Blocks for precision (best when Smart Environment embeds blocks). |
| [Results limit](https://smartconnections.app/smart-connections/settings/#connections-lists) | How many results are shown per note | Lower it for speed and less noise; raise it when you want more candidates. |
| [Connections List Component](https://smartconnections.app/smart-connections/settings/#display) (PRO) | Chooses the Pro renderer used by the main Connections list. **Version 4.0 (Graph + List)** adds the mini graph. | Use the ranked list as the primary workflow; select Graph + List only when the additional overview helps. |
| [Footer connections list component](https://smartconnections.app/smart-connections/settings/#footer-connections) | Component used only at the bottom of notes; **List only** keeps the Core and Pro footer compact. | In Pro, choose **Version 4.0 (Graph + List)** only when a footer mini graph is useful. |
| [Connections list item](https://smartconnections.app/smart-connections/settings/#display) (PRO) | Which metadata/snippet layout each result uses | When you want denser previews or different metadata emphasis. |
| [Connections sidebar location](https://smartconnections.app/smart-connections/settings/#display) | Which sidebar opens when you launch Connections | Use Right for a dedicated side panel; choose Left to keep it near file navigation. |
| [Render markdown](https://smartconnections.app/smart-connections/settings/#connections-lists) (PRO) | Whether snippets render markdown formatting | Turn on for rich previews; turn off to reduce rendering overhead. |
| [Include/Exclude path filters](https://smartconnections.app/smart-connections/settings/#filters) (PRO) | Restricts candidates by file path fragments | Focus results on a project area; hide archives/templates/noise. |
| [Frontmatter include/exclude filters](https://smartconnections.app/smart-connections/settings/#filters) (PRO) | Restricts candidates by YAML frontmatter key/value | Focus results on notes with matching metadata (ex. status:open); hide completed/archived work. |
| [Hide frontmatter blocks in results](https://smartconnections.app/smart-connections/settings/#filters) (PRO) | Hides YAML-derived blocks (frontmatter) from block results | Keep Blocks results focused on note content rather than metadata. |
| [Exclude inlinks/outlinks](https://smartconnections.app/smart-connections/settings/#filters) (PRO) | Removes notes already linked to/from the current note | When you want new link suggestions, not repeats of existing links. |
| [Scoring algorithm](https://smartconnections.app/smart-connections/settings/#score-algorithm) (PRO) | How initial similarity scores are computed | Start with Cosine Similarity; use feedback-based options if you want your signals to influence ordering. |
| [Ranking algorithm](https://smartconnections.app/smart-connections/settings/#ranking-algorithm) (PRO) | Reorders results after scoring (reranking) | When top results are close but ordering feels off; prioritize precision over speed. |
| [Inline connections score threshold](https://smartconnections.app/smart-connections/settings/#inline-connections) (PRO) | Minimum score required to show inline popovers | Raise it to reduce inline noise; lower it to see more matches while drafting. |
| [Footer connections](https://smartconnections.app/smart-connections/settings/#footer-connections) | Adds a Connections section at the bottom of notes | When you want a consistent place to browse related notes without opening the side panel, especially on mobile/no-sidebar workflows. |
| [Lookup results type/limit](https://smartconnections.app/smart-connections/settings/#lookup-lists) | Lookup granularity and number of results | Use Blocks for targeted answers; reduce limit for faster and simpler lookups where available. |
> [!Note]
> These settings control **what is shown** in Connections and Lookup. To control what is indexed/embedded, see **[Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page)**.
>
> Core includes the main ranked-list workflow, Sources vs Blocks, Connections results limit, sidebar location, Footer connections, Smart Lookup, Lookup result type, and baseline cosine similarity.
>
> Pro adds **Version 4.0 (Graph + List)**, inline connections, adjustable Lookup result limit, advanced result rendering, path/frontmatter filters, exclude inlinks/outlinks, hide frontmatter blocks, scoring algorithm selection, ranking/reranking, Bases integration, larger-vault performance workflows, and Smart Graph handoff.
## Where to find these settings in Obsidian
Open **Settings** -> **Community plugins** -> **Smart Connections** (or **Connections Pro**).
Many options are marked **PRO**. If you are not on Pro, you may see fewer controls than the screenshots on this page.
## Connections lists
These options change what appears in the main Connections list for the current note.
### Connection results type
Choose whether Connections returns:
- **Sources**: whole notes (coarser, usually faster)
- **Blocks**: sections/blocks within notes (more precise, usually heavier)
If you choose **Blocks**, you will get the best results when Smart Environment is configured to embed blocks (see [Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page)).
### Results limit
Controls how many Connections are displayed (default 20). Lower values reduce UI load; higher values show more candidates.
### Show full path (PRO)
When enabled, results include the folder path. This helps disambiguate similarly named notes and makes it easier to understand where a match lives in your vault.
### Render markdown (PRO)
When enabled, result snippets render markdown.
Turn this off if you:
- Prefer plain text snippets, or
- Want to avoid markdown rendering overhead in the list UI.
## Display
These settings control how the main Connections list is rendered and how list items look.
### Connections List Component (PRO)
Select the Pro component used to render the main Connections list.
**Version 4.0 (Graph + List)** renders the Connections mini graph above the ranked result list. Use the list to open and review a source before keeping a relationship.
Graph visibility is determined by the selected component. There is no separate **Show graph** setting.
This selector does not control Footer connections. Use **Footer connections list component** when you want to change only the note footer.
The current Graph + List renderer is a Pro capability. The main ranked-list workflow and Footer connections remain available in Core.
### Connections list item (PRO)
Select the list item renderer (template) used for each result.
This is mainly a UI preference: different item renderers may show different metadata, previews, or controls.
### Connections sidebar location
Choose which sidebar pane opens when you open the Connections view.
- **Right sidebar**: Keeps Connections in a dedicated panel.
- **Left sidebar**: Keeps Connections closer to file navigation.
- If a Connections tab is already open in the opposite sidebar, Smart Connections now reopens it in the configured location.
## Score algorithm
The score algorithm computes an initial similarity score between the current context and candidates.
### Scoring algorithm (PRO)
Select how items are scored against the current context.
Options shown in the UI include:
- **Cosine Similarity**
- **Similarity Adjusted by Feedback**
- **Similarity Weighted by Feedback**
- **Similarity Weighted by Key + Frontmatter**
The feedback-based options incorporate your feedback signals to influence future ordering. If you want a baseline, start with **Cosine Similarity**.
For full algorithm details and tuning guidance, see [Custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/).
### Cosine Similarity algorithm
Ranks by cosine similarity between the current note and candidates.
Score is a ranking signal, not a grade.
## Ranking algorithm
Ranking is a second step that can reorder results **after** the initial scoring step. This is where heavier algorithms (like reranking models) make sense, because they can be applied to a smaller set of top candidates.
### Ranking algorithm (PRO)
Enables a reranking strategy that modifies the order of results after initial scoring.
### Re-ranking model algorithm (PRO)
Uses a reranking model to adjust the order of connection results.
To use a reranking model, configure the default ranking model in Smart Environment Pro settings (see [Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page)).
## Connections filters
Filters hide items from the Connections results list.
### How filters match
Connections filters support two kinds of matchers:
- **Path filters** (Include filter, Exclude filter)
- Use **comma-separated** folder or file path fragments (example: `Projects/Clients`).
- Values are trimmed automatically.
- Matches use **case-sensitive substring** matching.
- **Frontmatter filters** (Frontmatter include filter, Frontmatter exclude filter)
- Use **one matcher per line**.
- Use `key` to match any value (example: `status`), or `key:value` to match a specific value (example: `status:open`).
- Matches use **case-insensitive** key and value matching.
### Result filters vs ingestion
Connections filters hide results **after** Smart Environment builds its dataset.
If you want to stop notes from being indexed/embedded at all, adjust Smart Environment include/exclude settings (Environment window or plugin settings). See [Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page).
### Precedence
- Entries in the **Exclude filter** always win when they match, even if the same path fragment appears in the **Include filter**.
- Entries in the **Frontmatter exclude filter** always win when they match, even if the same matcher appears in the **Frontmatter include filter**.
### Exclude inlinks (backlinks) (PRO)
Exclude notes that already link to the current note from the Connections results.
Use this when you want Connections to surface unlinked but relevant notes, not the ones already connected by backlinks.
### Exclude outlinks (PRO)
Exclude notes that are already linked from within the current note from appearing in the Connections results.
Use this to avoid re-suggesting notes you have already linked to.
### Include filter (PRO)
Comma-separated path fragments that must appear in the note path.
Use this to focus results on specific areas of your vault, like:
- `Daily/`
- `Projects/ClientA/`
- `People/`
> [!Note]
> This only affects the results list. Smart Environment may still embed matching notes unless they are excluded there.
### Exclude filter (PRO)
Comma-separated path fragments to omit from results.
Exclusions run before includes, so any matching fragment removes the note even if it also appears in Include filter.
If you want matching notes to not be embedded at all, use Smart Environment include/exclude settings (see [Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page)).
### Frontmatter include filter (PRO)
Newline-delimited frontmatter matchers that must be present in a candidate note's YAML frontmatter.
- One matcher per line
- `key` matches any value for that key
- `key:value` matches a specific value
- Key and value matching is case-insensitive
Example:
```text
status:open
type:meeting
project:client-a
```
Use this when you want Connections to stay inside a metadata slice of your vault (ex. only active projects).
### Frontmatter exclude filter (PRO)
Newline-delimited frontmatter matchers to remove from results.
Exclude entries take precedence over include entries.
Example:
```text
status:done
status:archived
area:private
```
Use this to hide completed, archived, or otherwise out-of-scope notes even if they would otherwise match by content.
### Hide frontmatter blocks in results (PRO)
When using **Blocks** results, enabling this hides blocks that come from note frontmatter (YAML), so the list focuses on note content rather than metadata.
## Inline connections
Inline connections show Connections at the block level inside the current note.
### Show inline connections (PRO)
When enabled, Smart Connections shows connections for each block within the note.
Hover the connections icon to see a list of connections for that block.
### Inline connections score threshold (PRO)
Minimum score (0 to 1) required before inline connections are displayed.
- Increase this to reduce noise (show fewer, stronger inline matches).
- Decrease this to see more matches.
### Skip code blocks (PRO)
When enabled, inline connections are not shown for blocks inside code blocks.
This is useful when:
- You keep lots of code snippets in notes, and
- You do not want inline connection UI on those sections.
## Footer connections
Footer connections show a Connections section at the bottom of each note.
> [!Note]
> Footer connections and the Footer list component selector are included in Smart Connections Core.
### Show footer connections
When enabled, Connections are displayed at the bottom of each note.
This is useful if you want a consistent place to browse related notes without opening the side panel. It is especially useful for mobile and no-panel workflows.
### Footer connections list component
Choose the component used to render Connections inside note footers.
- **List only** is the default. It shows the actionable result rows without a graph, keeping the note ending compact and well suited to mobile or no-sidebar workflows.
- **Version 4.0 (Graph + List)** (Pro) adds the Connections mini graph above the same result list. Choose it when spatial cues are worth the additional vertical space and rendering work.
This setting affects only Footer connections. It does not change the component used by the main Connections view or Connections codeblocks.
There is no separate **Show graph** toggle. Use **List only** for the compact Footer. In Pro, select **Version 4.0 (Graph + List)** to include the mini graph.
## Lookup lists
Lookup lists control what is shown in Lookup views (and other lookup-based lists).
### Lookup results type
Choose whether Lookup returns sources or blocks.
As with Connections results, block-level results are most useful when Smart Environment is configured to embed blocks (see [Smart Environment settings](https://smartconnections.app/smart-environment/settings/?utm_source=connections-settings-page)).
### Results limit (PRO)
Adjust the number of lookup results displayed.
## Related pages
- [Getting Started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/)
- [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/)
- [Footer connections](https://smartconnections.app/smart-connections/footer/)
- [Smart Lookup](https://smartconnections.app/smart-lookup/search/)
- [Inline connections](https://smartconnections.app/smart-connections/inline/)
- [Connections in Bases](https://smartconnections.app/smart-connections/bases/)
## FAQs
### What does "Sources vs Blocks" mean?
Sources returns whole notes for broader, faster context. Blocks returns smaller sections for finer precision and can increase storage/compute cost depending on settings. See [Connections settings](https://smartconnections.app/smart-connections/settings/), [Smart Lookup](https://smartconnections.app/smart-lookup/search/), and [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/).
### How do I tune relevance (reduce noise, focus on a project)?
Use limits, exclusions, and source type controls to narrow scope. For higher precision, test block-level results and rebuild after major config/model changes. See [Connections settings](https://smartconnections.app/smart-connections/settings/), [Smart Connections custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/), and [Smart Environment settings](https://smartconnections.app/smart-environment/settings/).
### How do I exclude folders/files from indexing?
Use Smart Environment exclusions for folders and noisy vault areas, then tune retrieval behavior in Connections settings. Start in [Smart Environment settings](https://smartconnections.app/smart-environment/settings/), then adjust [Connections settings](https://smartconnections.app/smart-connections/settings/) and verify via [Getting Started](https://smartconnections.app/smart-connections/getting-started/).
### Why does Footer show a list without a graph?
**List only** is the Footer default so the end of the note stays compact. Open the Footer connections settings and select **Version 4.0 (Graph + List)** (Pro) to include the mini graph.
### Does the Footer list component setting change the sidebar?
No. **Footer connections list component** changes only the component mounted at the bottom of notes. The main Connections view keeps its own **Connections List Component** setting.
---
## Milestones
canonical: https://smartconnections.app/smart-environment/milestones/
html_url: https://smartconnections.app/smart-environment/milestones/
markdown_url: https://smartconnections.app/smart-environment/milestones.md
llms_url: https://smartconnections.app/smart-environment/milestones/llms.txt
last_modified: 2026-08-26T15:52:34.662Z
usage_notes: |-
Use this page to answer questions about Milestones.
excerpt: |-
Smart Milestones Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Milestones documentation for current behavior. Start with Getting started with Smart Plugins for the shortest verified workflow. Smart Milestones are a guided, in-app checklist that helps you learn Smart Plugins by doing the smallest actions that unlock real value. Instead of…
suggested_links:
- title: Faq
url: https://smartconnections.app/smart-environment/faq/
- title: Settings
url: https://smartconnections.app/smart-environment/settings/
- title: Build Context In Obsidian Notes
url: https://smartconnections.app/build-context-in-obsidian-notes/
- title: Community Content
url: https://smartconnections.app/community-content/
- title: Faq
url: https://smartconnections.app/connect-pro/faq/
# Smart Milestones
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Milestones documentation](https://smartconnections.app/docs/plugins/#smart-plugins-milestones) for current behavior. Start with [Getting started with Smart Plugins](https://smartconnections.app/getting-started/) for the shortest verified workflow.
Smart Milestones are a guided, in-app checklist that helps you learn Smart Plugins by doing the smallest actions that unlock real value.
Instead of reading docs first, you can:
- Do one "first win" action
- Watch it get checked automatically
- Click any milestone to open the relevant docs when you want details
> [!TLDR] The Milestones loop
> 1. Open Milestones.
> 2. Complete the next unchecked item in the top group.
> 3. Use the feature once in a real note.
> 4. Repeat until you hit the workflow you care about (linking, lookup, context, chat).
## Where to find Milestones
Milestones live in the Smart Environment status bar menu.
1. Click the Smart Environment indicator in Obsidian's status bar.
2. Choose **Milestones**.
## How Milestones work
The Milestones modal is organized by feature area (Environment, Connections, Lookup, Context, Chat, and Pro).
What each line means:
- Each line is a "first time" action (open a view, run a command, drag a result, copy a context).
- A milestone checks itself the first time Smart Plugins detect you did it.
- Click any milestone to open the relevant docs (so you can go deeper only when you need to).
- Some milestones are labeled **PRO**. Those are optional and are there to show what is possible with Pro plugins enabled.
### How to read the UI quickly
- **Progress bar**: how many milestones are complete overall.
- **Group counters** (example: `3/8`): your progress in that feature area.
- **Check icon**: complete.
- **Empty circle**: not complete yet.
- **PRO badge**: requires Pro plugins.
## Your fastest first win
If you only do a few milestones, do these in order.
### 1) Finish setup (Environment)
Complete the Environment milestones so the rest of the features can work:
- "Initial vault import completed (all sources discovered)."
- "Initial embedding completed, you are ready to make connections!"
> [!Note]
> Connections and Lookup depend on embeddings. If embedding is still running, Milestones will reflect that.
### 2) Make your first meaningful link (Connections)
This is the shortest path from "I have notes" to "my vault is connected".
1. Open the Connections view.
2. Preview 1-2 results (Cmd/Ctrl hover).
3. Drag a result into your note to create a link.
Optional but useful:
- Open a random connection (it teaches you the "browse by meaning" behavior fast).
### 3) Use Lookup once (Lookup)
Use Lookup when you remember the idea, not the exact words.
1. Submit a lookup query (start a semantic search).
2. Open a result.
3. (Optional) Drag a Lookup result into a note to create a link.
### 4) Copy a tiny context pack (Context)
This is the bridge from "I found the right notes" to "I can paste grounded context into chat".
1. Create your first context.
2. Copy context to clipboard.
3. Open the Context Builder selector modal (so you know where saved packs live).
### 5) Close a loop with chat (Chat)
This turns "random chats" into trackable work.
1. Start a chat in a Smart Chat codeblock (opened the loop).
2. Mark the chat thread as done (closed the loop).
## Milestone notifications
When you complete a milestone, Smart Environment can show an achievement notification.
Use **View milestones** to jump back to the Milestones modal and pick the next action.
## Pick a path based on what you are trying to do
You do not need to complete everything. Use Milestones like a menu.
### If you want "related notes while writing"
Focus on **Connections**.
Next: try [Inline connections](https://smartconnections.app/smart-connections/inline/) so relevance shows up inside the editor.
### If you want "search by meaning"
Focus on **Lookup**.
Goal: run one query, open 1-2 results, link the best one.
### If you want "better AI answers with less copy/paste"
Focus on **Context**.
Next: use [Context Builder feature](https://smartconnections.app/smart-context/builder/) for reusable packs and [Context clipboard feature](https://smartconnections.app/smart-context/clipboard/) for one-off exports.
### If you want "threads that are organized inside your notes"
Focus on **Chat**.
Next: use [Chat codeblock](https://smartconnections.app/smart-chat/codeblock/) so each thread lives where the work lives.
## About Pro milestones
Pro milestones exist so you can:
- see what is possible,
- evaluate Pro by completing 1-2 high-signal actions,
- adopt only what actually improves your workflow.
If you are not using Pro, ignore PRO milestones and focus on completing the core milestones in Environment, Connections, Lookup, Context, and Chat.
## Troubleshooting
### "I did the thing, but it did not check"
Most milestones are detected the first time you perform an action. If one does not check:
- Do the action once more (slowly, in the same Obsidian window).
- Close and reopen the Milestones modal.
- Confirm prerequisites are complete (for example, embeddings must be ready before Connections and Lookup can work well).
### "Do I need to complete every milestone?"
No. Milestones are onboarding, not homework.
Complete the ones that match your workflow, and ignore the rest.
## Ready for your first win?
Open **Milestones** from the Smart Environment status bar, complete the next unchecked item, and follow the docs link to go deeper.
---
## Settings
canonical: https://smartconnections.app/smart-templates/settings/
html_url: https://smartconnections.app/smart-templates/settings/
markdown_url: https://smartconnections.app/smart-templates/settings.md
llms_url: https://smartconnections.app/smart-templates/settings/llms.txt
last_modified: 2026-08-26T15:51:07.216Z
usage_notes: |-
Use this page to answer questions about Settings.
excerpt: |-
Smart Templates settings Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Templates template-discovery documentation for current behavior. Start with Getting started with Smart Templates for the shortest verified workflow. Smart Templates settings control where reusable Markdown templates come from and how they are discovered. Open Settings…
suggested_links:
- title: Clipboard
url: https://smartconnections.app/smart-templates/clipboard/
- title: Commands
url: https://smartconnections.app/smart-templates/commands/
- title: Faq
url: https://smartconnections.app/smart-templates/faq/
- title: Generate
url: https://smartconnections.app/smart-templates/generate/
- title: Getting Started
url: https://smartconnections.app/smart-templates/getting-started/
# Smart Templates settings
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Templates template-discovery documentation](https://smartconnections.app/docs/templates/#smart-templates-template-discovery) for current behavior. Start with [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/) for the shortest verified workflow.
Smart Templates settings control where reusable Markdown templates come from and how they are discovered.
Open **Settings** -> **Community plugins** -> **Templates**.
## Templates folder
Choose one or more folders that contain reusable templates.
- Smart Templates searches the configured folders.
- When no folder is selected, the Obsidian Templates folder can be used as a fallback when that core plugin is configured.
- Review the suggestions in Template Context after changing this setting.
## Naming convention
Use a filename when one specific Markdown filename should count as a template outside the selected folders.
- `.md` is added automatically when missing.
- Matching the configured filename makes the note eligible as a template.
## Template headings
Choose headings when individual sections inside a Markdown note should act as templates.
- Matching heading blocks can appear as separate template suggestions.
- This lets one note contain several reusable template structures.
## Built-in templates
Built-in templates can still appear in Template Context even when no vault-backed template matches.
## Retired settings
Smart Templates 2.3 does not expose a Templates-specific **Generate model** or **Primary action** setting.
Those controls belonged to the retired v1 generation workflow. The current workflow ends at **Copy prompt**.
## Verify a settings change
1. Change one discovery setting.
2. Reopen **Smart Templates: Open template context**.
3. Confirm that the intended template or heading appears.
4. Select it, copy a small prompt, and inspect the pasted result.
## Related pages
- [Smart Templates](https://smartconnections.app/docs/templates/)
- [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/)
- [Smart Templates FAQ](https://smartconnections.app/smart-templates/faq/)
---
## Clipboard
canonical: https://smartconnections.app/smart-templates/clipboard/
html_url: https://smartconnections.app/smart-templates/clipboard/
markdown_url: https://smartconnections.app/smart-templates/clipboard.md
llms_url: https://smartconnections.app/smart-templates/clipboard/llms.txt
last_modified: 2026-08-26T15:49:25.511Z
usage_notes: |-
Use this page to answer questions about Clipboard.
excerpt: |-
Smart Templates Clipboard Deprecated page This page is retained temporarily and may contain outdated details. Use the current Smart Templates Clipboard documentation for current behavior. Start with Getting started with Smart Templates for the shortest verified workflow. Smart Templates Clipboard is the fastest way to turn your current note and your existing Markdown templates into a ready-to-run…
suggested_links:
- title: Commands
url: https://smartconnections.app/smart-templates/commands/
- title: Faq
url: https://smartconnections.app/smart-templates/faq/
- title: Generate
url: https://smartconnections.app/smart-templates/generate/
- title: Getting Started
url: https://smartconnections.app/smart-templates/getting-started/
- title: Modal
url: https://smartconnections.app/smart-templates/modal/
# Smart Templates Clipboard
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Smart Templates Clipboard documentation](https://smartconnections.app/docs/templates/#smart-templates-clipboard) for current behavior. Start with [Getting started with Smart Templates](https://smartconnections.app/smart-templates/getting-started/) for the shortest verified workflow.
Smart Templates Clipboard is the fastest way to turn your current note and your existing Markdown templates into a ready-to-run AI prompt.
It starts from context first, not template first.
1. Open the shared Template Context modal.
2. Start from the current note or selection.
3. Add more context if needed.
4. Select one or more templates.
5. Add optional instructions.
6. Click **Copy prompt**.
7. Paste into ChatGPT, Claude, Gemini, or any other chat UI.
## Who this helps
- You already have useful note structures in your vault.
- You want consistent outputs from web-based chat models.
- You do not want API setup just to reuse your templates.
- You want the template and the context to travel together.
## What makes this different
Most template workflows start with the template.
Smart Templates starts with the work in front of you.
That means:
- the current note becomes the starting context
- highlighted text can become the starting context instead
- template choice happens after you see and refine the context
## Multi-template support
Some jobs need more than one reusable structure.
Smart Templates lets you select multiple templates, then merges them in the order you selected them before copying the final prompt.
Use this when you want one output to satisfy several reusable constraints at once.
## Practical workflow
A simple loop:
1. Open a project note.
2. Select a structure template like summary, meeting output, or recommendation table.
3. Add 1 to 3 supporting notes to the context.
4. Add a short instruction like "focus on risks and next actions".
5. Copy the prompt.
6. Paste it into your chat tool.
Outcome: you stop rebuilding the same scaffolding every session.
## Related pages
- [Smart Templates getting started](https://smartconnections.app/smart-templates/getting-started/)
- [Smart Templates settings](https://smartconnections.app/smart-templates/settings/)
- [Smart Templates modal](https://smartconnections.app/smart-templates/modal/)
- [Retired Templates v1 generation workflow](https://smartconnections.app/docs/templates/#smart-templates-generate)
---
## Inline
canonical: https://smartconnections.app/smart-connections/inline/
html_url: https://smartconnections.app/smart-connections/inline/
markdown_url: https://smartconnections.app/smart-connections/inline.md
llms_url: https://smartconnections.app/smart-connections/inline/llms.txt
last_modified: 2026-08-26T14:54:39.594Z
usage_notes: |-
Use this page to answer questions about Inline.
excerpt: |-
Inline connections Deprecated page This page is retained temporarily and may contain outdated details. Use the current inline Connections documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. Inline connections bring related notes or blocks into the editor. Instead of switching to the Connections view, you can surface related…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
# Inline connections
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current inline Connections documentation](https://smartconnections.app/docs/connections/#smart-connections-inline) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
Inline connections bring related notes or blocks into the editor. Instead of switching to the Connections view, you can surface related material for a specific paragraph, heading, or sentence while you write.
> [!NOTE]
> Inline connections are part of Connections Pro. Use this page when Smart Connections already works for your vault and you want related material closer to the exact text you are editing.
>
> If you have not seen any useful Connections result yet, start with [Getting Started](https://smartconnections.app/smart-connections/getting-started/).
> [!TLDR]
> Inline connections answer: "what else is related to this block?" right where you are working.
> - Hover the inline Connections icon to see related matches.
> - Hold Cmd/Ctrl while hovering a match to open Obsidian Hover Preview.
> - Click a match to open its note.
> - Click the Connections + line numbers header to open the full list for that block in the Connections view.
> - Toggle inline connections only when you want them visible while editing.
## How it works
Inline connections score blocks from the current file against your vault based on your configured Connections settings. When a block has at least one match above your inline threshold, a small inline Connections icon appears at the end of that block.
![[Connections-Early-inline-decorator-2025-10-14.png]]
The icon is intentionally minimal: it is a fast signal that the current block has related context elsewhere.
The inline surface is most useful when the paragraph, heading, sentence, or block you are editing is the thing that should determine the results.
Use the main [Connections view](https://smartconnections.app/smart-connections/list-feature/) when the whole note should be the anchor or when you want a larger result list.
## Using inline connections
### 1. Hover to scan matches
Hover the Connections icon to open a popover list of related matches for that block.
Each result shows:
- the matched note or block
- a relative score for this inline result list
- a quick path to act, such as preview or open
Treat the score as a clue to inspect, not a decision. Preview the result before you act on it.
### 2. Preview before you switch context
Hold Cmd/Ctrl while hovering a connection to trigger Obsidian's native Hover Preview (above in pink).
This lets you inspect the match without leaving your current note.
Use preview when you want to answer:
- Is this actually relevant to this paragraph?
- Did I already explain this somewhere?
- Should this become a link, source, or follow-up note?
### 3. Open the full list for the block
Click the Connections + line numbers row in the popover to open all matches for that block in the main Connections view.
Use this when you want to:
- expand beyond the top few inline results
- switch from scanning to deeper exploration
- compare multiple candidates side-by-side in the Connections view
- copy links or send a result set to another workflow
### 4. Show inline connections only when you want them
If you prefer less visual noise while drafting, toggle inline connections on demand using the command palette, then assign that command to a hotkey if you use it often.
A practical flow:
1. Draft normally with inline connections off.
2. Toggle inline connections on when you want to add links, recover related context, or check repeated work.
3. Preview or open useful matches.
4. Toggle inline connections off when you return to focused writing.
## Workflow recipes
### Link while you draft
1. Write a paragraph.
2. Hover the inline icon.
3. Preview 1 to 2 promising matches.
4. Open the useful result or add it as a link.
5. Keep drafting.
Outcome: you build links as part of writing, not as a separate cleanup session.
### Ground a specific claim or paragraph
1. Find the sentence, paragraph, or heading that needs support.
2. Hover the inline icon for that block.
3. Preview related matches.
4. Add the useful source, link, or recovered decision to the current note.
Outcome: the current passage becomes better grounded in material already in your vault.
### Refactor long notes
1. Find a section that feels dense or disorganized.
2. Use inline connections on its blocks to locate related notes you already wrote.
3. Open the useful matches.
4. Split, merge, or extract the current section only when the match makes the next structure clearer.
Outcome: you reduce duplication and turn one long note into a more useful set of connected notes.
### Notice repeated work early
If you often restart research, inline connections can reveal "I already worked on this" at the moment you start writing.
Hover the inline icon and look for:
- a prior draft
- a similar concept explained elsewhere
- a related decision note or meeting note
Outcome: less rework, more reuse.
Use [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) when repeated notes or blocks need side-by-side review before you change them.
## Tuning inline visibility
Inline connections use Connections scoping controls, with extra controls for how much inline signal appears while editing.
Common controls include:
- Inline score threshold: how high a match must score before the icon appears
- Results limit: maximum items shown in the inline popover
- Include/exclude filters: restrict candidates by file path where available
For configuration details, see [inline connections settings](https://smartconnections.app/smart-connections/settings/#inline-connections).
> [!TIP]
> Inline connections only appear when a block has matches above your inline threshold.
>
> If normal Connections results work but inline icons do not appear, lower the inline score threshold and review inline settings.
>
> If no useful Connections results appear anywhere, use [Getting Started](https://smartconnections.app/smart-connections/getting-started/) to verify note eligibility and vault coverage first.
## When to use another surface
| When you need | Use |
| --- | --- |
| Related notes from the whole note | [Connections view](https://smartconnections.app/smart-connections/list-feature/) |
| Related notes at the bottom of the note | [Footer connections](https://smartconnections.app/smart-connections/footer/) |
| Relevance scoring inside a table or collection | [Connections in Bases](https://smartconnections.app/smart-connections/bases/) |
| Question-first retrieval | [Smart Lookup](https://smartconnections.app/smart-lookup/search/) |
| Duplicate or near-duplicate review | [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) |
| A reviewed AI context bundle | [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/) |
## Related pages
- [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/)
- [Inline connections settings](https://smartconnections.app/smart-connections/settings/#inline-connections)
- [Tune Smart Connections results, filters, and ranking](https://smartconnections.app/smart-connections/settings/)
- [Connections in Bases](https://smartconnections.app/smart-connections/bases/)
- [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/)
- [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/)
---
## Footer
canonical: https://smartconnections.app/smart-connections/footer/
html_url: https://smartconnections.app/smart-connections/footer/
markdown_url: https://smartconnections.app/smart-connections/footer.md
llms_url: https://smartconnections.app/smart-connections/footer/llms.txt
last_modified: 2026-08-25T22:22:59.086Z
usage_notes: |-
Use this page to answer questions about Footer.
excerpt: |-
Footer connections Deprecated page This page is retained temporarily and may contain outdated details. Use the current Footer Connections documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. Edition boundary Footer connections are available in Core and remain available when Connections Pro is active. The current Version 4.0…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
- title: Inline
url: https://smartconnections.app/smart-connections/inline/
# Footer connections
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Footer Connections documentation](https://smartconnections.app/docs/connections/#smart-connections-footer) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
> [!NOTE] Edition boundary
> Footer connections are available in Core and remain available when Connections Pro is active.
>
> The current **Version 4.0 (Graph + List)** renderer and its mini graph are Pro capabilities. **List only** is the compact Footer path.
Most orphan notes are created at the end of writing, not the beginning.
You finish the thought, close the note, and the chance to connect it is gone.
Footer connections solve that by placing a collapsible Connections panel at the bottom of the note you are already editing. When you reach the end of the note, related notes are waiting there: ready to scan, expand, open, or drag into the draft as links.
> [!TLDR]
> Footer connections are for the "finish this note well" moment.
> - Enable Footer connections.
> - Keep the default **List only** footer, or choose **Version 4.0 (Graph + List)** (Pro) in Connections settings.
> - Open a meaningful note and scroll to the bottom.
> - Review related notes before leaving the note.
> - Drag useful results into the note or open them for confirmation.
> - Collapse the panel when you want a cleaner note ending.
## Quick start
1. Enable Footer connections from [Connections settings](https://smartconnections.app/smart-connections/settings/#footer-connections), the command palette, or the ribbon icon.
2. Leave **Footer connections list component** set to **List only** for the compact default, or select **Version 4.0 (Graph + List)** (Pro) when you want a mini graph above the results.
3. Open a meaningful note and scroll to the bottom.
4. Use the footer to review related notes before you leave the page.
You can:
- click a result to open it
- expand a result to inspect more context
- drag a result into the editor to create an Obsidian link
- collapse the footer header when you want it out of the way
> [!TIP]
> The footer only appears once the end of the note is visible. That keeps it attached to the "done writing, now connect it" moment instead of adding noise higher in the note.
>
> If no useful Connections results appear anywhere in your vault, use [Getting Started](https://smartconnections.app/smart-connections/getting-started/) to verify note eligibility and vault coverage before tuning Footer.
> [!NOTE]
> Footer has its own list component selector. Changing it does not change the component used by the main Connections view or Connections codeblocks.
>
## Why Footer connections work
The main Connections view is useful when you want to explore a result list beside the note.
Footer connections are useful when the end of the note is the right review moment.
They show up when many notes either become connected or stay isolated. That makes them especially useful when you want to:
- link a draft before moving on
- recover a note you forgot existed
- add a quick Related or References section without opening a sidebar
- keep the workflow simple on mobile or in a no-sidebar setup
Use the main [Connections view](https://smartconnections.app/smart-connections/list-feature/) when you want broader exploration, pausing, copying links, or sending results to Smart Context.
Use Footer connections when the end of the note is the right review moment: finish writing, check related notes, add one useful link, and move on.
## What you can do from the footer
### Choose List only or Graph + List (Pro)
The **Footer connections list component** setting controls only the result component mounted at the bottom of notes.
- **List only** is the default. It keeps the note ending compact and puts the actionable result rows first.
- **Version 4.0 (Graph + List)** (Pro) adds the Connections mini graph above the same result list. Use it when spatial cues are worth the additional vertical space and rendering work.
Changing this setting rerenders Footer connections without changing the main Connections view or Connections codeblocks.
### Scan related notes before leaving the note
The footer keeps related notes close to the note that triggered them. This is useful when you want a fast answer to:
- "Did I already write about this?"
- "What should I link before I close this note?"
- "Which note would strengthen this draft most?"
### Expand before switching context
Expand a result when you want a little more context before opening it. This helps you confirm relevance without turning a small check into a bigger detour.
### Drag to create links on desktop
Drag a result from the footer into the editor to insert an Obsidian link directly into your note.
Use this when a suggested relationship should become part of the structure you author.
A useful pattern is:
```md
## Related
- [[...]]
## References
- [[...]]
```
### Open a result for closer review
Open a result when the preview is not enough and the useful next action is reading the source.
After reviewing it, return to the original note and add only the link, insight, source, or decision that helps the current work.
### Collapse the footer when you are done
Click the footer header to collapse or expand the panel. This keeps the feature available without forcing it to stay visually open all the time.
## Especially useful on mobile and no-sidebar workflows
Footer connections are often the most natural Connections workflow when a side panel is inconvenient.
Instead of opening and managing a sidebar, you stay in the note, scroll to the end, and connect ideas where the note already ends. That makes quick capture, review, and linking easier on a smaller screen or simpler layout.
The default **List only** component supports this compact workflow. Choose **Graph + List** (Pro) only when the additional visual overview helps more than the extra space costs.
> [!NOTE]
> Platform behavior can vary by Obsidian version, device, and current Smart Connections release. For workflow-critical mobile assumptions, review current mobile notes before publishing or documenting a team process.
## Workflow recipes
### 1) End-of-note linking pass
1. Finish writing the note.
2. Scroll to the footer.
3. Open or expand the top 1 to 3 promising results.
4. Add the useful link before leaving the note.
Outcome: fewer orphan notes and more chosen relationships in the notes you already write.
### 2) Mobile capture -> connect immediately
1. Capture or edit a note on mobile.
2. Scroll to the bottom.
3. Open or expand a useful related note.
4. Add the link while the connection is still obvious.
Outcome: quick notes become connected notes instead of loose fragments.
### 3) Final reference pass before sharing or publishing
1. Finish the draft.
2. Check the footer for adjacent notes, references, or prior versions.
3. Pull the useful matches into a Related or References section.
Outcome: the draft leaves the vault with the most useful nearby material attached.
### 4) No-sidebar review habit
1. Keep Footer connections enabled.
2. Write normally without managing a side panel.
3. Review the footer only when you reach the end of the note.
4. Add one useful result or collapse the panel and move on.
Outcome: related-note review becomes part of finishing work, not a separate maintenance ritual.
## Footer vs other Connections workflows
Use Footer connections when you want note-level suggestions at the end of the note. Its default **List only** component keeps this surface compact; select **Graph + List** (Pro) when you want the mini graph in the footer too.
Use the [Connections view](https://smartconnections.app/smart-connections/list-feature/) when you want broader exploration, pausing, copying links, or sending results to Smart Context.
Use [Inline connections](https://smartconnections.app/smart-connections/inline/) when you want related material beside a specific paragraph, heading, or sentence while editing.
Use [Connections settings](https://smartconnections.app/smart-connections/settings/#footer-connections) when you want to enable, disable, or choose the list component used by Footer connections.
## If Footer results do not help
| Symptom | Try first |
| --- | --- |
| No useful Connections results appear anywhere | Use [Getting Started](https://smartconnections.app/smart-connections/getting-started/) to verify note eligibility and vault coverage. |
| The footer does not appear | Scroll to the end of the note and confirm Footer connections are enabled in settings. |
| The footer has no graph | This is the default. Set **Footer connections list component** to **Version 4.0 (Graph + List)** when you want the mini graph. |
| The footer takes too much vertical space | Set **Footer connections list component** to **List only**. |
| Results are weak | Try a more meaningful note with enough text to act as the anchor. |
| The footer is visually distracting | Collapse the footer header until the next end-of-note review. |
| You want broader exploration | Use the [Connections view](https://smartconnections.app/smart-connections/list-feature/). |
| You want related material beside one paragraph | Use [Inline connections](https://smartconnections.app/smart-connections/inline/). |
## Related pages
- [Exploring the Connections view](https://smartconnections.app/smart-connections/list-feature/)
- [Connections settings: Footer connections](https://smartconnections.app/smart-connections/settings/#footer-connections)
- [Inline connections](https://smartconnections.app/smart-connections/inline/)
- [Getting Started](https://smartconnections.app/smart-connections/getting-started/)
---
## List Feature
canonical: https://smartconnections.app/smart-connections/list-feature/
html_url: https://smartconnections.app/smart-connections/list-feature/
markdown_url: https://smartconnections.app/smart-connections/list-feature.md
llms_url: https://smartconnections.app/smart-connections/list-feature/llms.txt
last_modified: 2026-08-25T22:14:08.490Z
usage_notes: |-
Use this page to answer questions about List Feature.
excerpt: |-
Exploring the Connections view Deprecated page This page is retained temporarily and may contain outdated details. Use the current Connections view documentation for current behavior. Start with Getting started with Smart Connections for the shortest verified workflow. The Connections view is the daily list for notes and blocks related to what you are looking at right now. Use this page after…
suggested_links:
- title: Bases
url: https://smartconnections.app/smart-connections/bases/
- title: Custom Algorithms
url: https://smartconnections.app/smart-connections/custom-algorithms/
- title: Faq
url: https://smartconnections.app/smart-connections/faq/
- title: Footer
url: https://smartconnections.app/smart-connections/footer/
- title: Getting Started
url: https://smartconnections.app/smart-connections/getting-started/
# Exploring the Connections view
> [!WARNING] Deprecated page
> This page is retained temporarily and may contain outdated details. Use the [current Connections view documentation](https://smartconnections.app/docs/connections/#smart-connections-view) for current behavior. Start with [Getting started with Smart Connections](https://smartconnections.app/smart-connections/getting-started/) for the shortest verified workflow.
The Connections view is the daily list for notes and blocks related to what you are looking at right now.
Use this page after Smart Connections is installed and you have already seen one useful related result. If you have not installed yet, or if your first run shows no useful results, start with [Getting Started](https://smartconnections.app/smart-connections/getting-started/).
Use the Connections view when the current note is the anchor:
> What else in my vault is related to what I am looking at right now?
This guide shows how to preview results, act on useful matches, keep the right anchor steady, reduce noise, copy links, and send selected results into the next workflow.
> [!NOTE] Watch related notes appear while the note changes
> [](https://youtu.be/_i3577ti8jg?t=316)
>
> Callum types into a new note and the related notes begin to appear. That is the current-note anchor in action: more meaningful text gives Connections better material to retrieve from.
## When to use the Connections view
Use the Connections view when you are already working in a note and want related material to stay visible beside the work.
It helps with:
- rediscovering forgotten work at the moment it becomes useful
- linking related notes without browsing folders or hunting search results
- noticing prior related material before you rewrite it
- turning useful suggestions into chosen links, sources, or follow-up actions
> [!NOTE]
> You do not need perfect folders, tags, or prior links. The note in view supplies the starting point, and you choose which results become part of your work.
## Quick start: preview one result, then use it
Use the Connections list when the note in front of you is the starting point.
The goal is not to browse a list. The goal is to recover one useful idea, source, prior decision, or relationship and use it in the note you are already working on.
1. Open a note.
2. Open connections view
3. Scan the top results.
4. Gain an insight that helps progress the current note.
You know it worked when a connection reveals something relevant to the current note: an idea, source, prior decision, useful relationship, or material you can reuse.
You know it became useful when the current note changes because of that result.
> [!TIP]
> Do not start with an empty test note.
>
> Open a project note, meeting note, draft, research note, or decision note with enough text to represent what you are thinking about. Connections uses the note in view to find related notes.
A simple first artifact is:
```md
## Related
- [[Useful related note]]
```
If you have not yet seen any useful result from Smart Connections, review [Getting Started](https://smartconnections.app/smart-connections/getting-started/) before continuing.
## Connections vs Lookup vs exact search
Use the surface that matches the anchor.
| What starts the task | Use | Why it fits |
| --- | --- | --- |
| A note you are already working in | Connections view | The current note determines the results. |
| A question or idea you can type | [Smart Lookup](https://smartconnections.app/smart-lookup/search/) | The written query determines the results. |
| Exact phrase, regex, filename, heading, or tag | Obsidian search | You need lexical matching. |
Use this rule:
```md
Current note -> Connections.
Question -> Lookup.
Exact phrase -> Obsidian search.
```
Connections is not a replacement for exact search.
It is for the moment when related material should surface from the note you are already working in.
Related notes are not automatically duplicates. Use [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) when repeated material needs review before changing notes.
## Core moves after one useful result
After one result helps the current note, these moves make the list useful in daily work.
Keep the loop simple:
```md
current note -> related result -> preview -> useful action
```
### Preview without leaving the note
Preview helps you understand why a result appeared before you switch context.
Use either preview path:
1. Expand a result in the list to inspect more content.
2. On desktop, hold Cmd/Ctrl while hovering a result to use Obsidian Hover Preview.
Preview is not a trust ceremony and the score is not a grade. Treat each result as a lead. Keep the result only if it helps the current work.
Use preview when you want to check:
- whether the result contains useful prior reasoning
- whether the relationship is meaningful, not merely similar
- whether the note should be opened, linked, ignored, or reviewed later
### Drag one useful result into the next workflow
Drag a verified result into the editor to create an Obsidian link, or move it directly into Smart Chat, an existing named-context row, Smart Graph, or the live Connections view.
Use this when you want to:
- preserve a useful relationship in the current note
- ground the current Smart Chat response
- curate the result into a reusable saved context
- compose or extend a graph scope
- make the result the fixed Connections target
A good first link pass is small:
```md
## Related
- [[One useful related note]]
```
Do not turn the list into clutter. Add the result because it helps the current note, not because it appeared near the top.
### Hold one anchor steady with Pause
Connections updates as the active note changes.
Use Pause when you want to keep one note as the reference point while you inspect or open results.
This is useful when you want to:
- open a result without changing the list immediately
- compare several results against the same note
- keep context stable while reviewing related material
If results do not change when you switch notes, check whether Pause is active. Switch back to Play when you want the list to follow the active note again.
### Copy as a list of links
Copy results as a clean list of Obsidian links when the result set itself should become an artifact.
Use this for:
- Related notes sections in drafts
- meaning-ranked reading trails
- project hub updates
- reference lists for review
### Send results to Smart Context
Send results to Smart Context when the useful matches should become a reviewable context bundle for AI work.
A practical flow:
1. Pause the list on the note you are working from.
2. Send results to Smart Context.
3. Remove anything noisy.
4. Reorder what matters.
5. Copy the final bundle into the AI workflow that needs it.
Related:
- [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/)
- [Smart Context Builder](https://smartconnections.app/smart-context/builder/)
## Understanding the list
### Play/Pause updates
Play/Pause controls whether the list updates as you change notes.
Use Pause when:
- you are writing and want stable context
- you want to browse while keeping one anchor note as your reference point
- you are comparing several candidate notes against the same anchor
Use Play when you want the list to refresh as the active note changes.
If the list feels stuck, make sure Play is active.
### Drop one item anywhere in the live view
Drop one indexed note or block anywhere inside the rendered Connections view to pause live following and make that item the current target. The target includes the top bar, graph, result list, expanded content, and background.
Folders and multi-item drops are rejected because the view can display only one target at a time.
### Score
Score is a ranking signal, not a grade.
Higher scores generally mean "more related" inside the current result list, but the number is a clue to inspect, not a guarantee.
Practical rules:
1. Compare scores within the same list, not across different notes.
2. Score ranges vary by vault content and model.
3. Preview results before acting on them.
4. If results feel broad or noisy, tune Sources vs Blocks, limits, and filters in [Connections settings](https://smartconnections.app/smart-connections/settings/).
For scoring and ranking controls, see [Connections settings](https://smartconnections.app/smart-connections/settings/) and [Custom algorithms](https://smartconnections.app/smart-connections/custom-algorithms/).
### Expand/collapse
Expand a result to preview content without opening the note.
Use it to scan faster and reduce context switching.
### Click behavior
Clicking a result follows Obsidian's default link behavior:
- Click: open in the current pane.
- Cmd/Ctrl + click: open in a new tab.
- Cmd/Ctrl + Alt + click: open in a new pane or split.
## Managing noise: Hide and Pin
As your vault grows, some connections will be technically related but not useful right now.
### Hide
Hide removes a noisy result from the list.
Use Hide when:
- a note is a frequent false positive
- a template, index, or archive note pollutes results
- the topic is adjacent but not relevant to the current task
### Pin
Pin keeps an item visible as a stable reference.
Pinning is useful when:
- one reference should stay visible while you browse
- you want a small set of important notes to remain easy to reach
- a result is useful enough that you expect to return to it during the session
> [!NOTE]
> Some Connections Pro scoring options can use pinned and hidden signals when you choose a feedback-aware algorithm.
## List controls
The menu contains controls that turn scanning into action.
### Refresh
Refresh recomputes results for the current note.
Use it when:
- you made major edits and want updated relationships
- you changed settings and want to re-run the list
- the list feels stale for the note in view
### Send to Smart Context
Send results to Smart Context when the current list should become a reviewable bundle for delegating work to AI.
Use it after you have identified useful matches, not as a substitute for choosing what matters.
Learn more:
- [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/)
- [Smart Context Builder](https://smartconnections.app/smart-context/builder/)
### Copy as a list of links
Copy results as a simple list of Obsidian links.
Use it to:
- create a Related notes section
- build a reading trail
- paste ranked references into a project hub
- preserve the current result set before changing notes
### Quick controls cheat sheet
| Control | What it does | When to use it |
| --- | --- | --- |
| Play/Pause | Stop or resume automatic updates | Keep one anchor note steady while you browse, or resume updates for the active note. |
| Refresh | Recompute results | After major edits or settings changes. |
| Copy links | Copy ranked links | Add a clean related list to a note, hub, or trail. |
| Send to Context | Send results to Smart Context | Build a reviewable AI-context bundle from useful matches. |
| Hide/Pin | Remove noise or keep essentials visible | Reduce clutter or keep key references near the work. |
## Related surfaces
Use these when your starting point changes.
| When you need | Use |
| --- | --- |
| Related notes at the bottom of the note | [Footer connections](https://smartconnections.app/smart-connections/footer/) |
| Related material beside a specific paragraph while editing | [Inline connections](https://smartconnections.app/smart-connections/inline/) |
| Relevance scoring inside a table or collection | [Connections in Bases](https://smartconnections.app/smart-connections/bases/) |
| Topic shape, clusters, or neighborhoods | [Smart Graph](https://smartconnections.app/smart-graph/) |
| A reviewable bundle for AI work | [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/) |
Each linked page owns its own setup, screenshots, controls, and workflow recipes.
## Common Connections view workflows
### Writing: ground a draft in your existing notes
1. Open the draft.
2. Scan the top Connections results.
3. Preview 1-2 promising matches.
4. Drag the best result into a Related or References section.
5. Keep writing with the recovered material in view.
Outcome: the draft uses what your vault already knows instead of starting from memory alone.
### Research: build a meaning-ranked reading trail
1. Open the note that best represents the topic or question.
2. Pause the Connections list on that note.
3. Copy results as links.
4. Paste them into a Reading trail note.
5. Review from the top, removing anything that does not help.
Outcome: the current note becomes the anchor for a reading sequence without manual folder browsing.
### Review: recover prior reasoning before making a decision
1. Open the decision, project, or meeting note.
2. Scan for earlier notes that look related.
3. Preview the strongest result.
4. Add the recovered decision, source, or constraint to the current note.
5. Open the source only if the preview is not enough.
Outcome: earlier reasoning returns before you repeat the same work.
### AI work: assemble grounded context fast
1. Open the note that defines the current assignment.
2. Pause the Connections list.
3. Send results to Smart Context.
4. Remove noise and keep only useful matches.
5. Ask for a specific outcome using the reviewed context.
Outcome: the AI workflow starts from selected vault context instead of a broad prompt.
## When results feel stale or stuck
Use the smallest fix that matches what happened.
| What happened | Try first |
| --- | --- |
| Results are empty on a tiny or test note | Open a richer note with meaningful text. |
| Results are not updating when you change notes | Make sure Play is active, not Pause. |
| Relationships feel stale after major edits | Use Refresh in the Connections view. |
| A specific expected note never appears | Check whether the note or folder is excluded in Smart Environment. |
| Results are broad but at least one is useful | Keep using the useful result before tuning settings. |
| Results are repeatedly noisy after the basic workflow works | Then use limits, Sources vs Blocks, filters, hide, or Pro ranking controls. |
| You have a question rather than a note to start from | Use [Smart Lookup](https://smartconnections.app/smart-lookup/search/) instead. |
| You know the exact word, filename, heading, tag, syntax, or regex | Use Obsidian search. |
| Similar results look like repeated work | Use [Smart Dedupe](https://smartconnections.app/smart-dedupe/getting-started/) to review likely duplicates before changing notes. |
| You need the shape of a topic, not a current-note list | Use [Smart Graph](https://smartconnections.app/smart-graph/). |
If one meaningful note unexpectedly has no results, inspect the active note and confirm it can be embedded.
If results are absent across many notes, check vault preparation and embedding coverage in [Smart Environment settings](https://smartconnections.app/smart-environment/settings/) or [Connections settings](https://smartconnections.app/smart-connections/settings/).
## Local retrieval details
For how local embeddings, provider-backed workflows, and privacy boundaries work, see:
- [WHY local embeddings](https://smartconnections.app/smart-connections/why-local-embeddings/)
- [Smart Environment settings](https://smartconnections.app/smart-environment/settings/)
## FAQs
See [Connections FAQs](https://smartconnections.app/smart-connections/faq/).
## Related pages
- [Getting Started](https://smartconnections.app/smart-connections/getting-started/)
- [Smart Connections overview](https://smartconnections.app/smart-connections/)
- [Smart Lookup](https://smartconnections.app/smart-lookup/search/)
- [Connections settings](https://smartconnections.app/smart-connections/settings/)
- [Footer connections](https://smartconnections.app/smart-connections/footer/)
- [Inline connections](https://smartconnections.app/smart-connections/inline/)
- [Connections in Bases](https://smartconnections.app/smart-connections/bases/)
- [Smart Context Clipboard](https://smartconnections.app/smart-context/clipboard/)
---
## Plugins
canonical: https://smartconnections.app/docs/plugins/
html_url: https://smartconnections.app/docs/plugins/
markdown_url: https://smartconnections.app/docs/plugins.md
llms_url: https://smartconnections.app/docs/plugins/llms.txt
last_modified: 2026-08-25T18:26:47.141Z
usage_notes: |-
Use this page to answer questions about Plugins.
excerpt: |-
Smart Plugins Use the Smart Plugins Store to install Core and Pro plugins. Complete the next lifecycle action. If the workflow uses indexed sources, prepare Smart Environment. Confirm one plugin result. New to Smart Plugins? Start with Getting Started with Smart Plugins for the first workflow. Use this page for exact Store and Environment recovery. In this guide Install and update Smart Plugins…
suggested_links:
- title: Chat
url: https://smartconnections.app/docs/chat/
- title: Connect Pro
url: https://smartconnections.app/docs/connect-pro/
- title: Connections
url: https://smartconnections.app/docs/connections/
- title: Context
url: https://smartconnections.app/docs/context/
- title: Dedupe
url: https://smartconnections.app/docs/dedupe/
# Smart Plugins
Use the Smart Plugins Store to install Core and Pro plugins. Complete the next lifecycle action. If the workflow uses indexed sources, prepare Smart Environment. Confirm one plugin result.
> [!TIP] New to Smart Plugins?
> Start with [Getting Started with Smart Plugins](https://smartconnections.app/smart-plugins/getting-started/) for the first workflow. Use this page for exact Store and Environment recovery.
## In this guide
- [Install and update Smart Plugins in Obsidian](#smart-plugins-store)
- [Prepare Smart Environment for a plugin workflow](#smart-environment-runtime)
- [Track Smart Plugin progress with Smart Milestones](#smart-plugins-milestones)
## Install and update Smart Plugins in Obsidian
The Smart Plugins Store is the in-app catalog for Core and Pro Smart Plugins. Core and Pro identify product tracks. They do not tell you whether a plugin is installed, loaded, configured, or producing a useful result.
### Open the Smart Plugins Store
Select **Browse Smart Plugins** from one of these locations:
- Smart Environment settings
- a Smart Plugin settings page
- the Smart Environment status menu
- the Command Palette
### Complete the shortest install path
1. Find the plugin.
2. Make sure that its **Core** or **Pro** label is correct.
3. Select the action shown by the row: **Install**, **Install Core**, or **Install Pro**.
4. If the row shows **Enable**, select it.
5. Complete any exact **Reload required...** action shown by the Store.
6. When the row shows **Active**, open its settings if configuration is required.
7. If the workflow uses indexed sources, prepare Smart Environment.
8. If the workflow uses indexed sources, continue when the status view shows **Smart Environment ready** or **Ready**.
9. Provider codeblocks can skip the Environment steps.
10. Open the plugin.
11. Make sure that the plugin gives one result you can examine.
The numbers explain the Store boundaries in the image:
1. **Core** and **Pro** identify product tracks, not workflow success.
2. **Included in Pro** describes how a Core capability relates to Pro. It is not an activation state.
3. **Active**, **Open settings**, and the enabled toggle show that the Pro plugin is loaded.
4. **Installed** with **Enable** means the Core plugin files exist but the plugin is disabled.
Use the Core or Pro label to select the edition. Follow the action shown on the live row. **Active** confirms that a plugin is loaded. It does not confirm that the workflow is configured or working. Store versions and update actions change with releases. Use the current row, not a remembered label.
### Read the account row
The account row can show these exact states and actions:
| Account state | Next action |
| --- | --- |
| **Checking session...** | Wait for the session check to finish. |
| **Connect account** | Select **Login**, or **Copy link to login instead**. |
| **Session needs refresh** | Select **Refresh**. Use **Logout** when the saved session should be removed instead. |
| **All-access subscription expired** | Select **Get Pro**, **Update subscription**, or **Refresh**, as appropriate. |
| Signed in | Pro install and update access follows the current account access. **Logout** remains available. |
Signed-out users can browse the catalog and install eligible Core plugins. Pro installation and updates require current Pro access.
### Browse the catalog sections
The Store shows the main catalog and a separate **Experimental** section when experimental plugins are available.
### Interpret Store lifecycle states
Follow the exact label shown. The Store does not use a standalone **Reload** state or a visible **Configured** state.
| Row label or action | Meaning |
| --- | --- |
| **Install** / **Install Core** | Install the eligible Core package. |
| **Install Pro** | Install the Pro package when the account has access. |
| **Requires Pro** | Current access does not permit that Pro install. |
| **Included in Pro** | The capability is included with the Pro track shown by the Store. |
| **Core installed** | The Core package is installed while the grouped row also describes a Pro option. |
| **Installed** + **Enable** | Files are installed, but the plugin is disabled. Select **Enable**. |
| **Enabled** | The plugin is enabled in Obsidian but is not reported as loaded and active in the current session. |
| **Active** | The plugin is loaded. **Open settings** and the enabled toggle are available when applicable. |
| **Update to vX** | A newer eligible package is available. Apply the update. Then follow any reload-required label. |
| **Reload required to activate vX** | Reload Obsidian so the installed version becomes active. |
| **Reload required to disable** | Reload Obsidian to finish disabling the plugin. |
| **Reload required for Smart Environment** | Reload Obsidian so the required Smart Environment state can take effect. |
An enabled toggle labeled **Enabled** describes the toggle state. It is not the same as the **Enable** action for an installed but disabled plugin.
### Separate loaded, configured, ready, and working
Use four checks:
1. **Active**: the plugin code is loaded.
2. **Configured**: required model, provider, or plugin settings are complete. This is a workflow check, not a Store label.
3. **Smart Environment ready** or **Ready**: required only when the workflow uses shared indexed sources. It means that Environment work is loaded and idle.
4. **Working**: the plugin produces one result you can examine.
Do not stop at **Installed**, **Enabled**, or **Active** when the desired outcome is a usable workflow.
### Update an installed plugin
When the Store shows **Update to vX**, select it. Complete the exact reload-required action that appears. Make sure that the row returns to **Active**. Test the plugin result again.
Smart Environment v3 is a coordinated suite upgrade. Update every installed Smart Plugin before you restart Obsidian. Mixed v2 and v3 environments are blocked so a partial upgrade cannot appear healthy.
### Open settings and release information
Use **Open settings** for the active plugin. Use the row's details or release action for the documentation or release page available to that edition.
### Recover when a Store action does not complete
| Symptom | Recovery |
| --- | --- |
| Pro install or update is unavailable | Resolve **Connect account**, **Session needs refresh**, or **All-access subscription expired** in the account row. |
| Row shows **Installed** | Select **Enable**. |
| Row shows a reload-required label | Select that exact action and let Obsidian reload. |
| Row shows **Enabled**, not **Active** | Make sure that no reload-required action remains. Reload Obsidian. Reopen the Store. |
| **Events & notifications** says a plugin is waiting to load and shows **Restart Obsidian** | Select **Restart Obsidian**. Reopen the Store. Make sure that the plugin row is **Active** before testing its workflow. |
| Row shows **Active**, but the workflow fails | Complete plugin configuration. For an indexed-source workflow, make sure that Smart Environment is **Ready**. Make sure that the sources are eligible. Test one result. |
| Store says Smart Environment is required | Complete **Reload required for Smart Environment**. Then use the Environment recovery below. |
## Prepare Smart Environment for a plugin workflow
Smart Environment supplies source discovery, import, embedding, model settings, status, events, and diagnostics used by Smart Plugins.
### Read the Environment status
| Status | Meaning or next action |
| --- | --- |
| **Smart Environment not loaded** / **Idle** | Select **Load Smart Environment**. |
| **Loading Smart Environment** | Wait for loading to finish. |
| **Importing** / **Re-importing** | Source preparation is running. |
| **Embedding** | Embedding preparation is running. |
| **Embedding paused** | Select **Resume embedding** when preparation should continue. |
| **Queued re-import work** | Select **Run re-import** when the queued work should start now. |
| **Smart Environment ready** / **Ready** | The Environment is loaded and idle. Test the plugin result. |
What the numbers identify:
1. **Smart Environment ready** means Environment is loaded with no active import or embedding work.
2. **Open events feed** and **Environment stats** open retained events and source-coverage diagnostics.
3. **Browse Smart Plugins** returns to installation and activation controls.
The status view also provides **Pause embedding**, **Open events feed**, **Environment stats**, and **Browse Smart Plugins**.
### Load Smart Environment from settings
When Environment settings show the pre-load state, select **Load Smart Environment**. **Show loading status** opens the current loading view.
### Configure the Environment
Use the exact settings that match the task:
- **Default embedding model** chooses the fallback embedding model.
- **Default chat model** chooses the fallback Chat model when no other model is specified.
- **Default ranking model** chooses the fallback ranking model used by applicable features, including Connections re-ranking.
- **Test model** checks the selected model.
- **Re-index embeddings** starts embedding preparation for the current model and policy.
- **Embed blocks** makes eligible note sections available as smaller retrieval units.
- **Re-import wait time** delays automatic re-import after file activity.
- **Manage excluded folders** controls folder exclusions.
- **Manage excluded files** is available in Pro for file-level exclusions.
- **View all exclusions** reviews the combined exclusion set.
- **Reset data** opens an explicit source-data clearing confirmation. Choose **Cancel** unless a rebuild is intended. Choose **Re-import** to begin the rebuild. The confirmation does not establish deletion of Markdown note files.
Installed model lists expose **+ New** and row controls such as **Edit**, **Test**, and **Current**. **Current** means selected. It does not prove that a test succeeded, authorization is valid, or the model will produce useful output.
What the numbers identify:
1. **Manage excluded folders** changes folder-level source boundaries.
2. **Manage excluded files** adds Pro file-level exclusions.
3. **View all exclusions** reviews the combined exclusion set.
4. **Embed blocks** and minimum-length controls change which source and block items are eligible.
What the numbers identify:
1. The selected built-in embedding model for this configuration.
2. The exact available model labels, including choices marked experimental.
3. **Re-index embeddings** and **Test model** as separate actions after model selection.
### Export Environment data
Choose **Export data** from the Environment status-bar menu. The export produces one JSON file.
1. Select **Sources**, **Blocks**, or both. At least one collection is required.
2. Enable **Include embedding vectors** only when downstream work needs the current vector in each item's `vec` field. Vectors can make the export much larger.
3. Choose **Export**. The dialog states that the JSON file will be saved in the vault root.
4. Monitor the visible **Sources** and **Blocks** progress phases and the output filename.
5. On completion, make sure that the filename and 100 percent state appear, then open the file in the vault root. **Export again** starts another export.
A failed run shows `Export failed: ...` and **Retry export**. Do not treat an in-progress or failed state as success, and do not rely on the dialog alone when the JSON file must be delivered.
### Inspect one source
Keep the intended note active. Choose **Inspect active note** from the Environment status-bar menu. Source inspector can show the source identity and path, properties and links, block coverage, source data, and indexed block excerpts.
Use its status filters to narrow the diagnostic view:
- **All** shows the complete current diagnostic set.
- **Needs embedding** and **Current** separate pending from current block state.
- **Skipped** shows omitted block rows and their visible diagnostics.
- **Unexpected** can expose a retained row labeled **Unexpected vector**.
These states help locate a boundary. They do not establish root cause, repair, re-import success, or retrieval quality.
### Recover by symptom
| Symptom | Recovery |
| --- | --- |
| One active note is missing | Keep the note active. Select **Inspect active note** from the status menu. |
| Several notes are missing | Select **Show stats** from the status menu or **Environment stats** from the status view. Examine eligibility. Examine current embeddings. Examine exclusions. |
| Import or embedding appears stuck | Select **Open events feed**. Examine retained events before you reset data. |
| Embedding is paused | Select **Resume embedding**. |
| Re-import work is queued | Select **Run re-import**. |
| Plugin is **Active** but returns no usable result | Make sure that configuration is complete. For an indexed-source workflow, make sure that Environment is **Ready**. Make sure that the relevant sources are eligible. Examine the product result. |
What the numbers identify:
1. Total indexed items across Smart Sources and Smart Blocks.
2. Work still needed plus any unexpected embeddings.
3. Smart Sources eligibility and current coverage.
4. Smart Blocks eligibility and current coverage.
## Track Smart Plugin progress with Smart Milestones
Smart Milestones records supported first-use actions and can suggest the next feature to try. Confirm usefulness in the plugin's own workflow rather than treating a checked milestone as the result itself.
What the numbers identify:
1. Overall detected progress across the visible milestone set.
2. A completed Environment group.
3. A partially completed Connections group.
4. A partially completed Lookup group.
Open the related plugin. Examine the actual result. Milestone progress does not grade result quality.
### Open Smart Milestones
Select **Milestones** from the Smart Environment status menu.
### Review milestone progress
The modal groups milestones by plugin and workflow. Complete only the groups that support the workflows you are adopting.
A checked row means the detector recorded its action, event, or installed-plugin state. Select a milestone to open its documentation when detailed steps or recovery are needed.
### Understand optional Pro milestones
Pro milestones are labeled separately and require the related Pro plugin, prerequisites, and account access.
### Achievement notifications
A completed milestone can show an achievement notification after Obsidian becomes idle. Notifications appear one at a time while the app is visible.
### Troubleshoot an unchecked milestone
1. Confirm that the action exists in the installed plugin release.
2. Confirm its prerequisites, including source import and embedding readiness when that milestone uses indexed sources.
3. Repeat the action in the same Obsidian window.
4. Reopen Smart Milestones.
5. Inspect the feature result even if the milestone becomes checked.
## Related documentation
- [Use the Connections view](https://smartconnections.app/docs/connections/#find-related-notes-in-obsidian-with-smart-connections)
- [Run Smart Lookup](https://smartconnections.app/docs/lookup/#search-obsidian-notes-by-meaning-with-smart-lookup)
- [Open Smart Context Builder](https://smartconnections.app/docs/context/#build-reusable-ai-context-bundles-in-obsidian)
- [Save a provider thread](https://smartconnections.app/docs/chat/#embed-and-organize-ai-chat-threads-in-obsidian)
---