Outline Wiki MCP Server
Used for text chunking in the smart features implementation to process and embed wiki documents for semantic search and retrieval.
Used for AI-powered smart features including text embeddings (text-embedding-3-small) for vector search and GPT-4o-mini for RAG-based Q&A, document summarization, tag suggestions, and diagram generation.
Enables LLMs to interact with Outline wiki through structured API calls, providing document management, search, collections, comments, and AI-powered features including RAG-based Q&A, summarization, and semantic document discovery.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Outline Wiki MCP Server search for documents about quarterly planning"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Outline Wiki MCP Server
A Model Context Protocol (MCP) server that enables LLMs to interact with Outline wiki through structured API calls. This server provides document management, search, collections, comments, and AI-powered smart features including RAG-based Q&A.
Supported Clients
Client | Tools | Resources | Prompts |
✅ | ✅ | ✅ | |
✅ | ✅ | ✅ | |
✅ | ✅ | ✅ | |
✅ | - | ✅ | |
✅ | - | - | |
✅ | - | - |
Getting Started
Requirements
Node.js 18.0.0 or higher
Outline instance with API access
(Optional) OpenAI API key for smart features
Getting Your Outline API Token
Log in to your Outline instance
Go to Settings → API
Click Create API Key
Copy the generated token (starts with
ol_api_)
Installation
Add to your Claude Desktop configuration:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.jsonLinux:
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx"
}
}
}
}Run the following command:
claude mcp add outline -e OUTLINE_URL=https://your-outline-instance.com -e OUTLINE_API_TOKEN=ol_api_xxxxxxxxxxxxx -- npx -y outline-mcp-serverOr add to ~/.claude.json (global) or .mcp.json (project-local):
{
"mcpServers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx"
}
}
}
}Note: The
~/.claude/settings.jsonfile is ignored for MCP servers. Use~/.claude.jsonor.mcp.jsoninstead.
Add to your VS Code settings (.vscode/mcp.json):
{
"servers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx"
}
}
}
}Add to Cursor MCP settings (~/.cursor/mcp.json):
{
"mcpServers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx"
}
}
}
}Add to Windsurf MCP settings (~/.codeium/windsurf/mcp_config.json):
{
"mcpServers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx"
}
}
}
}ChatGPT supports MCP through its desktop app. Add the server in Settings → MCP Servers with:
Command:
npxArguments:
-y outline-mcp-serverEnvironment variables as shown above
Configuration
Environment Variables
Variable | Description | Required | Default |
| Your Outline instance URL | Yes |
|
| Your Outline API token | Yes | - |
| Enable read-only mode | No |
|
| Disable delete operations | No |
|
| API retry attempts | No |
|
| Retry delay (ms) | No |
|
| Enable AI features | No |
|
| OpenAI API key | No* | - |
* Required when ENABLE_SMART_FEATURES=true
Smart Features Configuration
To enable AI-powered features (RAG Q&A, summarization, etc.), add these to your config:
{
"mcpServers": {
"outline": {
"command": "npx",
"args": ["-y", "outline-mcp-server"],
"env": {
"OUTLINE_URL": "https://your-outline-instance.com",
"OUTLINE_API_TOKEN": "ol_api_xxxxxxxxxxxxx",
"ENABLE_SMART_FEATURES": "true",
"OPENAI_API_KEY": "sk-xxxxxxxxxxxxx"
}
}
}
}Tools
Search & Discovery
Tool | Description |
| Search documents by keyword with pagination |
| Find document ID by title |
| Get all collections |
| Get document hierarchy in a collection |
| Get recently modified documents |
Document Operations
Tool | Description |
| Get full document content by ID |
| Export document in Markdown |
| Create a new document |
| Update document (supports append) |
| Move document to another location |
Document Lifecycle
Tool | Description |
| Archive a document |
| Restore archived document |
| Delete document (soft/permanent) |
| Restore from trash |
| List archived documents |
| List trashed documents |
Comments & Collaboration
Tool | Description |
| Add comment (supports replies) |
| Get document comments |
| Get specific comment |
| Find linking documents |
Collection Management
Tool | Description |
| Create collection |
| Update collection |
| Delete collection |
| Export collection |
| Export all collections |
Batch Operations
Tool | Description |
| Create multiple documents |
| Update multiple documents |
| Move multiple documents |
| Archive multiple documents |
| Delete multiple documents |
Smart Features (AI-Powered)
Requires ENABLE_SMART_FEATURES=true and OPENAI_API_KEY.
Tool | Description |
| Check status and indexed count |
| Sync docs to vector database |
| RAG-based Q&A on wiki content |
| Generate AI summary |
| AI-suggested tags |
| Find semantically related docs |
| Generate Mermaid diagrams |
Smart Features Usage
# 1. First, sync your wiki documents
sync_knowledge
# 2. Ask questions about your wiki
ask_wiki: "What is our deployment process?"
# 3. Summarize long documents
summarize_document: { documentId: "doc-id", language: "Korean" }
# 4. Find related content
find_related: { documentId: "doc-id", limit: 5 }Technology Stack
Component | Technology |
Vector Database | LanceDB (embedded) |
Embeddings | OpenAI text-embedding-3-small |
LLM | GPT-4o-mini |
Text Chunking | LangChain |
Safety Features
Read-Only Mode
READ_ONLY=trueRestricts to read operations only: search, get, export, list operations, and all smart features.
Disable Delete
DISABLE_DELETE=trueBlocks: delete_document, delete_collection, batch_delete_documents
Development
# Clone repository
git clone https://github.com/huiseo/outline-wiki-mcp.git
cd outline-wiki-mcp
# Install dependencies
npm install
# Build
npm run build
# Run tests
npm test
# Type check
npm run typecheckLicense
MIT License - see LICENSE for details.
Links
Available Tools
37 toolsadd_commentC
Add a comment to a document. Supports replies.
Inputs: documentId (required), text (required), parentCommentId.
archive_documentD
Archive a document.
Inputs: documentId (required).
ask_wikiC
Ask a question and get an answer based on wiki content using RAG.
Inputs: question (required).
batch_archive_documentsC
Archive multiple documents at once.
Inputs: documentIds (required).
batch_create_documentsC
Create multiple documents at once.
Inputs: documents (required).
batch_delete_documentsC
Delete multiple documents at once.
Inputs: documentIds (required), permanent.
batch_move_documentsC
Move multiple documents at once.
Inputs: documentIds (required), collectionId, parentDocumentId.
batch_update_documentsC
Update multiple documents at once.
Inputs: updates (required).
create_collectionC
Create a new collection.
Inputs: name (required), description, color.
create_documentD
Create a new document.
Inputs: title (required), collectionId (required), text, parentDocumentId, +1 more.
delete_collectionB
Delete a collection. All documents in the collection will also be deleted.
Inputs: collectionId (required).
delete_documentB
Delete a document. If permanent=true, permanently delete; otherwise move to trash.
Inputs: documentId (required), permanent.
export_all_collectionsC
Export all collections.
Inputs: format.
export_collectionC
Export a collection.
Inputs: collectionId (required), format.
export_documentC
Export document in Markdown format.
Inputs: documentId (required).
generate_diagramC
Generate a Mermaid diagram from a text description.
Inputs: description (required).
get_collection_structureC
Get document hierarchy within a collection.
Inputs: collectionId (required).
get_commentC
Get details of a specific comment.
Inputs: commentId (required).
get_documentC
Get full document content by document ID.
Inputs: documentId (required).
get_document_backlinksC
Find other documents linking to this document.
Inputs: documentId (required).
get_document_id_from_titleC
Find document ID by title.
Inputs: query (required), collectionId.
list_archived_documentsC
Get list of archived documents.
Inputs: limit.
list_collectionsC
Get list of all collections.
Inputs: no named parameters.
list_document_commentsC
Get list of comments on a document.
Inputs: documentId (required), limit, offset.
list_recent_documentsC
Get list of recently modified documents.
Inputs: limit.
list_trashC
Get list of documents in trash.
Inputs: limit.
move_documentC
Move document to another collection or under a parent document.
Inputs: documentId (required), collectionId, parentDocumentId.
restore_documentC
Restore a document from trash.
Inputs: documentId (required).
search_documentsC
Search documents by keyword. Supports pagination.
Inputs: query (required), collectionId, limit, offset.
smart_statusB
Check if smart features are enabled and get index statistics.
Inputs: no named parameters.
suggest_tagsC
Get AI-suggested tags for a document based on its content.
Inputs: documentId (required).
summarize_documentC
Generate an AI-powered summary of a document.
Inputs: documentId (required), language.
sync_knowledgeB
Sync documents to vector store for AI-powered search. Run this before using ask_wiki.
Inputs: collectionId.
unarchive_documentB
Restore an archived document.
Inputs: documentId (required).
update_collectionC
Update collection information.
Inputs: collectionId (required), name, description, color.
update_documentC
Update an existing document. Can append content with append mode.
Inputs: documentId (required), title, text, append.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 37 tools
Most tools have distinct purposes, but some overlap exists between search_documents, find_related, and ask_wiki (all involve finding content). Batch operations are clearly differentiated from single operations, and archive/unarchive vs. trash/restore are well separated.
All tools follow a consistent snake_case verb_noun pattern (e.g., create_document, list_collections, update_document). The naming is highly predictable and uniform across all 37 tools.
37 tools is excessive for a wiki server, making the interface heavy and potentially overwhelming. While the domain is rich, many tools could be consolidated (e.g., batch operations as parameters) or omitted without losing core functionality.
The toolset provides comprehensive coverage for wiki management, including full CRUD for documents and collections, archiving/trash workflows, export capabilities, AI features (summarize, suggest_tags, ask_wiki), and advanced operations like backlinks and related documents.
Related MCP Connectors
Ingest, manage, and retrieve documents for RAG-powered AI applications
Personal wiki and memory layer for AI assistants. Persistent, structured memory across sessions.
Governed, auditable knowledge your team curates for its AI assistants, self-hostable
Persistent docs and memory for AI agents — read, write, organize & search a shared workspace.