Wiki.js MCP Server

What is Wiki.js MCP Server?

A Model Context Protocol (MCP) server that integrates AI assistants with Wiki.js to create, manage, and synchronize hierarchical documentation across multiple repositories. Designed for organizations with large-scale documentation needs.

How to use Wiki.js MCP Server?

Clone the repository, configure environment variables in .env, start Wiki.js via Docker (docker-compose -f docker.yml up -d), and run ./setup.sh to install dependencies. Then either start the MCP server with ./start-server.sh or configure it in Cursor by adding the command to ~/.cursor/mcp.json.

Key features of Wiki.js MCP Server

• Hierarchical documentation with nested page creation
• GraphQL API integration for Wiki.js v2+
• File-to-page mapping and code structure analysis
• Docker one-command deployment with PostgreSQL
• Bulk operations and change tracking
• Enterprise scalability for hundreds of repos

Use cases of Wiki.js MCP Server

• Automatically create repository documentation structures for new projects
• Keep documentation in sync with source code changes
• Clean up deprecated documentation hierarchies safely
• Manage multi-repository documentation in a single Wiki.js instance
• Auto‑organize project files into categorized documentation

FAQ from Wiki.js MCP Server

What dependencies are needed to run the server?

Python 3.x with FastMCP, httpx, SQLAlchemy, Pydantic, and tenacity. Docker is recommended for running the Wiki.js instance itself.

How does authentication work?

You can use a JWT token (API key from Wiki.js admin panel) or provide username/password credentials in the .env file.

Where does the server store its local data?

It uses a local SQLite database (wikijs_mappings.db) for file‑to‑page mappings and repository context information.

How do I integrate the server with Cursor?

Add the start-server.sh command to your ~/.cursor/mcp.json file, then click the refresh icon in Cursor to load all 21 tools.

Are there any known limits on repositories or pages?

No explicit limits are mentioned; the system is designed for enterprise scalability handling hundreds of repositories and thousands of files.