Medullar MCP Server

What is Medullar MCP Server?

Medullar MCP Server is an MCP protocol server that retrieves a user’s Medullar Space data. It is built for developers and users who need to access their curated Medullar spaces through an MCP‑compatible client (e.g., Claude Desktop).

How to use Medullar MCP Server?

Install Python 3.10+, the uv package manager, and dependencies. Create a .env file with your MEDULLAR_JWT token. Run in development mode with mcp dev server.py or install in Claude Desktop with mcp install server.py. For custom deployments, use python server.py or mcp run server.py.

Key features of Medullar MCP Server

• Retrieves the user’s Medullar spaces via MCP protocol.
• Requires a Medullar JWT for authentication.
• Returns space name, UUID, and context.
• Supports .env file, shell variables, or mcp install -v.
• Built with Python 3.10+ and the uv package manager.

Use cases of Medullar MCP Server

• Query your curated Medullar spaces from an MCP‑capable UI.
• Integrate space data into AI assistants like Claude Desktop.
• Automate retrieval of space metadata for custom workflows.

FAQ from Medullar MCP Server

What prerequisites are needed?

Python 3.10 or higher, the uv package manager, and a valid Medullar JWT token.

How do I provide my JWT token?

Store it in a .env file as MEDULLAR_JWT=your_token, or set it as an environment variable in your shell, or pass it via the -v flag with mcp install.

What data can I retrieve with the server?

The only endpoint is get_medullar_spaces, which returns a list of your spaces with their name, uuid, and context.

How do I run the server in production?

Install it for Claude Desktop with mcp install server.py, or run directly with python server.py or mcp run server.py.