memos
A FastMCP-based MCP (Model Context Protocol) server implementation for Memos, allowing AI assistants to interact with the Memos note-taking system through the MCP protocol.
claude mcp add --transport stdio mylxsw-memos-mcp-server uvx memos-mcp-server \ --env MEMOS_API_KEY="your-api-key-here" \ --env MEMOS_SERVER_URL="https://memos.example.com" \ --env MEMOS_MCP_AUTH_TOKEN="optional-token-if-needed"
How to use
This MCP server provides an interface to interact with a Memos note-taking backend using the Model Context Protocol. It exposes tools to create, read, update, delete, and list memos, with optional authentication support. You can query for memo details, fetch recent memos, and manage memo content with Markdown support. The server communicates over HTTP at the MCP endpoint and is designed to integrate with AI assistants and plugins that speak MCP-compatible commands.
Key capabilities include: create_memo to add new memos with Markdown content, update_memo to modify content, state, visibility, or pin status, delete_memo to remove a memo, and info at memos://{memo_resource_name}/info to fetch detailed memo metadata. You can also retrieve recent memos via memos://recent_memos. The protocol and endpoints are designed to be consumed by clients that understand FastMCP-style commands and responses.
How to install
Prerequisites:
- Python 3.13 or newer
- uv (FastMCP/uv-based tooling) installed
- Access to a running Memos server (MEMOS_SERVER_URL) and an API key (MEMOS_API_KEY)
Install and run locally:
-
Clone the repository: git clone <repository-url> cd memos-mcp
-
Install dependencies and start the server with uv: uv sync uv run main.py
-
Configure environment variables (example): export MEMOS_SERVER_URL="https://your-memos-server.com" export MEMOS_API_KEY="your-api-key"
Optional: authentication for MCP clients
export MEMOS_MCP_AUTH_TOKEN="your-auth-token"
-
Verify the service starts and the MCP endpoint is available at: http://localhost:8000/mcp
Docker (optional):
- Docker Compose:
- Set MEMOS_SERVER_URL and MEMOS_API_KEY in a .env file or your environment
- docker-compose up -d
- Docker run:
docker build -t memos-mcp .
docker run -p 8000:8000
-e MEMOS_SERVER_URL="https://your-memos-server.com"
-e MEMOS_API_KEY="your-api-key"
memos-mcp
Additional notes
Tips and common considerations:
- Ensure MEMOS_SERVER_URL and MEMOS_API_KEY are correctly set; the MCP server will reuse these to talk to the Memos backend.
- If you enable MCP authentication, provide MEMOS_MCP_AUTH_TOKEN to restrict access.
- The MCP endpoint is typically exposed at http://localhost:8000/mcp with HTTP protocol; adjust firewall and port mappings as needed in your environment.
- If you encounter network or authentication issues, verify that the Memos API is reachable from the MCP server and that the API key is valid.
- When using Docker, remember to expose port 8000 if you want to access the MCP endpoint from outside the container host.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP