mcp -memos

A MCP(Model Context Protocol) server for Memos.

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio leslieleung-mcp-server-memos uvx memos-mcp-server \
  --env MEMOS_URL="https://memos.example.com" \
  --env DEFAULT_TAG="#mcp" \
  --env MEMOS_API_KEY="your_api_key"

How to use

This MCP server provides an integration with Memos to enable querying and creating memos through the MCP protocol. The server exposes two tools: search_memos, which searches memos by keyword or tags, and create_memo, which creates a new memo entry in your Memos instance. To use it, run the MCP server via your preferred runtime (via uvx as shown in the configuration) and connect to it through your MCP client. The DEFAULT_TAG environment variable can be used to tag new memos by default, and MEMOS_URL / MEMOS_API_KEY are used to authenticate and route requests to your Memos instance. When you invoke search_memos, you can pass a keyword and optional filters; create_memo requires at least a title or content and will attach it to your Memos workspace.

How to install

Prerequisites:

Ensure you have a runtime that supports uvx (Python/uv) installed on your system.
Access to the Memos instance with a valid MEMOS_API_KEY.

Install steps:

Install uvx (Python-based MCP runtime) if not already installed.
Ensure Node is not required for this server (per the provided config); this server uses uvx to run the memos-mcp-server binary.
Configure environment variables as described in the mcp_config (MEMOS_URL, MEMOS_API_KEY, DEFAULT_TAG).
Start the MCP server with the provided configuration, for example by launching the uvx runtime with the memos-mcp-server package:

uvx memos-mcp-server
Verify the server is listening and reachable by your MCP client and test the search_memos and create_memo tools.

Additional notes

Tips:

MEMOS_URL must point to your Memos instance API endpoint. Ensure the API key has permission to read/write memos.
DEFAULT_TAG is prepended to new memos created via create_memo if no tag is provided.
If you encounter authentication errors, double-check MEMOS_API_KEY and MEMOS_URL accessibility from the host running the MCP server.
The server exposes only the memo-focused tools (search_memos and create_memo); other MCP features may not be available unless extended.
When deploying in production, consider securing environment variables and rotating MEMOS_API_KEY periodically.

Related MCP Servers

web-eval-agent

1.2k

An MCP server that autonomously evaluates web applications.

mcp-neo4j

911

Neo4j Labs Model Context Protocol servers

Gitingest

136

mcp server for gitingest

zotero

132

Model Context Protocol (MCP) server for the Zotero API, in Python

fhir

FHIR MCP Server – helping you expose any FHIR Server or API as a MCP Server.

unitree-go2

The Unitree Go2 MCP Server is a server built on the MCP that enables users to control the Unitree Go2 robot using natural language commands interpreted by a LLM.