mcp-fathom-analytics

This MCP server exposes a set of Fathom Analytics tools via the Model Context Protocol so an AI assistant can retrieve account details, manage sites, fetch events, and generate analytics. The available tools include: get-account for account details, list-sites to enumerate your Fathom sites, list-events to fetch events for a specific site, get-aggregation to build custom analytics reports with flexible filters and groupings, and get-current-visitors to see real-time visitor activity. To use the server with an MCP client (for example Claude Desktop or another MCP client), point the client at the server configuration and authorize with your Fathom API key. The server acts as a bridge between the MCP client and the Fathom Analytics API via the included SDK.

Prerequisites:

• Node.js (recommended LTS) and npm installed on your machine
• Internet access to fetch the MCP server package from npm

Installation steps:

1. Ensure you have Node.js and npm installed. Verify with: node -v npm -v

2. Run the MCP server via npx (no local install required): npx -y mcp-fathom-analytics

3. Configure environment variables (example):

   • FATHOM_API_KEY: your Fathom Analytics API key

4. If you prefer to run via npm/yarn script, you can install the package and then run the server through your own script as needed (see the package documentation for any additional run parameters).

Notes and tips:

• You must provide a valid Fathom API key in FATHOM_API_KEY to access the API.
• The server uses the @mackenly/fathom-api SDK; ensure your API key has the necessary permissions for the actions you intend to perform (account, sites, events, aggregations, current_visitors).
• The MCP client should be configured to call the server’s exposed tools using the MCP specification (mcpServers structure). Use the example as a starting point and adapt the server name or keys as needed.
• If you encounter rate limits or API errors, verify your API key and check the Fathom API status. You can also adjust aggregation queries to fetch smaller time ranges when testing.
• This project is published to npm as an npx script, so running via npx is the recommended quick-start path.