mirror of
https://github.com/simstudioai/sim.git
synced 2026-04-06 03:00:16 -04:00
76486ebcc8
* feat(knowledge): add v1 knowledge base API, Obsidian/Evernote connectors, and docs
- Add v1 REST API for knowledge bases (CRUD, document management, vector search)
- Add Obsidian and Evernote knowledge base connectors
- Add file type validation to v1 file and document upload endpoints
- Update OpenAPI spec with knowledge base endpoints and schemas
- Add connectors documentation page
- Apply query hook formatting improvements
* fix(knowledge): address PR review feedback
- Remove validateFileType from v1/files route (general file upload, not document-only)
- Reject tag filters when searching multiple KBs (tag defs are KB-specific)
- Cache tag definitions to avoid duplicate getDocumentTagDefinitions call
- Fix Obsidian connector silent empty results when syncContext is undefined
* improvement(connectors): add syncContext to getDocument, clean up caching
- Update docs to say 20+ connectors
- Add syncContext param to ConnectorConfig.getDocument interface
- Use syncContext in Evernote getDocument to cache tag/notebook maps
- Replace index-based cache check with Map keyed by KB ID in search route
* fix(knowledge): address second round of PR review feedback
- Fix Zod .default('text') overriding tag definition's actual fieldType
- Fix encodeURIComponent breaking multi-level folder paths in Obsidian
- Use 413 instead of 400 for file-too-large in document upload
- Add knowledge-bases to API reference docs navigation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(knowledge): prevent cross-workspace KB access in search
Filter accessible KBs by matching workspaceId from the request,
preventing users from querying KBs in other workspaces they have
access to but didn't specify.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(knowledge): audit resourceId, SSRF protection, recursion depth limit
- Fix recordAudit using knowledgeBaseId instead of newDocument.id
- Add SSRF validation to Obsidian connector (reject private/loopback URLs)
- Add max recursion depth (20) to listVaultFiles
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(obsidian): remove SSRF check that blocks localhost usage
The Obsidian connector is designed to connect to the Local REST API
plugin running on localhost (127.0.0.1:27124). The SSRF check was
incorrectly blocking this primary use case.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
144 lines
7.0 KiB
Plaintext
144 lines
7.0 KiB
Plaintext
---
|
|
title: Connectors
|
|
description: Automatically sync documents from external sources into your knowledge base
|
|
---
|
|
|
|
import { Callout } from 'fumadocs-ui/components/callout'
|
|
import { Step, Steps } from 'fumadocs-ui/components/steps'
|
|
|
|
Connectors let you pull documents directly from external services into your knowledge base. Instead of manually uploading files, a connector continuously syncs content from sources like Notion, Google Drive, GitHub, Slack, and more — keeping your knowledge base up to date automatically.
|
|
|
|
## Available Connectors
|
|
|
|
Sim ships with 20+ built-in connectors spanning productivity tools, cloud storage, development platforms, and more.
|
|
|
|
| Category | Connectors |
|
|
|----------|-----------|
|
|
| **Productivity** | Notion, Confluence, Asana, Linear, Jira |
|
|
| **Cloud Storage** | Google Drive, Dropbox, OneDrive, SharePoint |
|
|
| **Documents** | Google Docs, WordPress, Webflow |
|
|
| **Development** | GitHub |
|
|
| **Communication** | Slack |
|
|
| **CRM** | HubSpot, Salesforce |
|
|
| **Data** | Airtable |
|
|
| **Note-taking** | Evernote, Obsidian |
|
|
| **Meetings** | Fireflies |
|
|
|
|
## Adding a Connector
|
|
|
|
<Steps>
|
|
<Step>
|
|
|
|
### Select a source
|
|
|
|
Open a knowledge base and click **Add Connector**. You'll see the full list of available connectors — pick the service you want to sync from.
|
|
|
|
</Step>
|
|
<Step>
|
|
|
|
### Authenticate
|
|
|
|
Most connectors use **OAuth** — select an existing credential from the dropdown, or click **Connect new account** to authorize through the service's login flow. Tokens are refreshed automatically, so you won't need to re-authenticate unless you revoke access.
|
|
|
|
A few connectors (Evernote, Obsidian, Fireflies) use **API keys** instead. Paste your key or developer token directly, and it will be stored securely.
|
|
|
|
<Callout type="info">
|
|
If you rotate an API key in the external service, you'll need to update it in Sim as well. OAuth tokens are refreshed automatically, but API keys are not.
|
|
</Callout>
|
|
|
|
</Step>
|
|
<Step>
|
|
|
|
### Configure
|
|
|
|
Each connector has its own configuration fields that control what gets synced. Some examples:
|
|
|
|
- **Notion**: Choose between syncing an entire workspace, a specific database, or a single page tree
|
|
- **GitHub**: Specify a repository, branch, and optional file extension filter
|
|
- **Confluence**: Enter your Atlassian domain and optionally filter by space key or content type
|
|
- **Obsidian**: Provide your vault URL and optionally restrict to a folder path
|
|
|
|
All configuration is validated when you save — if a repository doesn't exist or a domain is unreachable, you'll get an immediate error.
|
|
|
|
</Step>
|
|
<Step>
|
|
|
|
### Choose sync frequency
|
|
|
|
Select how often the connector should re-sync:
|
|
|
|
| Frequency | Description |
|
|
|-----------|-------------|
|
|
| Every hour | Best for fast-moving sources |
|
|
| Every 6 hours | Good balance for most use cases |
|
|
| **Daily** (default) | Suitable for content that changes infrequently |
|
|
| Weekly | For stable, rarely-updated sources |
|
|
| Manual only | Sync only when you trigger it |
|
|
|
|
</Step>
|
|
<Step>
|
|
|
|
### Configure metadata tags (optional)
|
|
|
|
If the connector supports metadata tags, you'll see checkboxes for each tag type (e.g., Labels, Last Modified, Notebook). All are enabled by default — uncheck any you don't need.
|
|
|
|
See the [Metadata Tags](#metadata-tags) section below for details.
|
|
|
|
</Step>
|
|
<Step>
|
|
|
|
### Connect & Sync
|
|
|
|
Click **Connect & Sync** to save the connector and trigger the first sync immediately. Documents will begin appearing in your knowledge base as they are processed.
|
|
|
|
</Step>
|
|
</Steps>
|
|
|
|
## How Syncing Works
|
|
|
|
On each sync, the connector fetches documents from the external service and compares them against what's already in your knowledge base. Only documents that have actually changed are reprocessed — new content is added, updated content is re-chunked and re-embedded, and documents that no longer exist in the source are removed.
|
|
|
|
This means syncing is efficient even for large document sets. A connector with thousands of documents will only do meaningful work when something changes.
|
|
|
|
### Handling Failures
|
|
|
|
If a single document fails to fetch (e.g., due to a permission issue or timeout), the sync continues with the remaining documents. The failed document will be retried on the next sync cycle.
|
|
|
|
If an entire sync fails (e.g., the service is down or credentials expired), the connector automatically backs off and retries later. The backoff resets as soon as a sync succeeds.
|
|
|
|
## Metadata Tags
|
|
|
|
Connectors can automatically populate [tags](/docs/knowledgebase/tags) with metadata from the source, letting you filter documents in the Knowledge block based on information from the external service.
|
|
|
|
For example, a Notion connector might tag documents with their **Labels**, **Last Modified** date, and **Created** date. A GitHub connector might tag documents with their **Repository** and **File Path**. This metadata becomes available for [tag-based filtering](/docs/knowledgebase/tags) in your workflows.
|
|
|
|
### Opting Out
|
|
|
|
You can disable specific metadata tags during connector setup. Disabled tags won't be populated, leaving those tag slots available for other connectors or manual tagging.
|
|
|
|
<Callout type="info">
|
|
Tag slots are shared across all documents in a knowledge base. If you have multiple connectors, each one's metadata tags draw from the same pool of available slots.
|
|
</Callout>
|
|
|
|
## Excluding Documents
|
|
|
|
You can manually exclude specific documents from a connector's sync. Excluded documents are skipped on every subsequent sync, even if they change in the source. This is useful for filtering out templates, drafts, or other content you don't want in your knowledge base.
|
|
|
|
## Source Links
|
|
|
|
Every synced document retains a link back to the original in the external service. This lets you trace any knowledge base document to its source — whether that's a Notion page, a GitHub file, a Confluence article, or a Slack conversation.
|
|
|
|
## Multiple Connectors
|
|
|
|
You can add multiple connectors to a single knowledge base. For example, you might sync internal documentation from Confluence alongside code from GitHub and meeting notes from Fireflies — all searchable together through the Knowledge block.
|
|
|
|
Each connector manages its own documents independently. Metadata tag slots are shared across the knowledge base, so keep an eye on slot usage if you're combining several connectors that each populate tags.
|
|
|
|
## Common Use Cases
|
|
|
|
- **Internal knowledge base**: Sync your team's Notion workspace and Confluence spaces so AI agents can answer questions about internal processes, policies, and documentation
|
|
- **Customer support**: Connect HubSpot or Salesforce alongside your help docs from WordPress or Google Docs to give support agents full context on customers and product information
|
|
- **Engineering assistant**: Sync a GitHub repository and Jira or Linear issues so an AI agent can reference code, specs, and ticket history when answering developer questions
|
|
- **Meeting intelligence**: Pull in Fireflies transcripts alongside Slack conversations to build a searchable archive of decisions and discussions
|
|
- **Research and notes**: Sync Evernote notebooks or an Obsidian vault to make your personal notes available to AI workflows
|