medullar-mcp-server

medullar-mcp-server

0

The Medullar MCP Server is designed to retrieve data from Medullar Spaces using the MCP protocol. It supports various installation and execution modes to cater to both development and production environments.

Medullar MCP Server

A server for retrieving Medullar Space data using the MCP protocol.

Prerequisites

  • Python 3.10 or higher
  • uv package manager (recommended)
  • A Medullar JWT token

Setup

  1. Install dependencies using uv:
brew install uv
uv pip install -r requirements.txt
  1. Create a .env file in the project root with your Medullar JWT:
MEDULLAR_JWT=your_jwt_token_here

Running Your Server

Development Mode

The fastest way to test and debug your server is with the MCP Inspector:

mcp dev server.py

Production Installation

Once your server is ready, install it in Claude Desktop:

mcp install server.py

Direct Execution

For advanced scenarios like custom deployments:

python server.py
# or
mcp run server.py

Environment Variables

The server requires the following environment variables:

  • MEDULLAR_JWT: Your Medullar authentication token

These can be provided either through:

  • A .env file in the project root
  • Environment variables in your shell
  • The mcp install command with the -v flag

API Endpoints

The server provides the following endpoints:

get_medullar_spaces

Retrieves the user's Medullar spaces, which contain their curated data.

Returns a list of spaces in the format:

[
    {
        "name": "Space Name",
        "uuid": "Space unique identification",
        "context": "Space context"
    }
]

Changelog

See for a list of changes and version history.