mcp-limitless

This MCP server bridges your Limitless Pendant data to AI tools via the Limitless API, exposing a set of MCP tools that let AI assistants query and retrieve Lifelog entries. It provides endpoints to list lifelogs by date or range, fetch a specific lifelog by ID, list recent lifelogs, and search through recent lifelogs by text. You can connect this server to MCP clients such as Claude, Windsurf, Cursor, or ChatWise, enabling chat-based workflows that pull actionable insights from your Pendant data. The tools are designed to work with the Limitless API and require a valid LIMITLESS_API_KEY to access your data.

To use, configure your MCP client to launch the built Node.js server with the environment variable LIMITLESS_API_KEY set. The client will spawn the server process as a stdio server, and you can invoke the exposed tools like limitless_get_lifelog_by_id, limitless_list_lifelogs_by_date, limitless_list_lifelogs_by_range, limitless_list_recent_lifelogs, and limitless_search_lifelogs through your client interface.

Prerequisites:

• Node.js v18 or later
• npm or yarn
• A Limitless account and API key
• A Limitless Pendant (required for data)

Installation steps:

1. Clone this repository or download the project files.
2. Navigate to the project directory: cd mcp-limitless-server
3. Install dependencies: npm install
4. Build the server: npm run build
5. Obtain the absolute path to the built server script (the dist/server.js file). You will reference this path in your MCP client configuration, for example /Users/you/project/mcp-limitless-server/dist/server.js.

Notes:

• Do not run npm start directly; configure your MCP client to spawn the server as a stdio server.
• In your MCP client configuration, ensure LIMITLESS_API_KEY is provided as an environment variable.

Tips and considerations:

• The Limitless API is in beta and may experience instability or rate limits. Large requests can time out; the server applies a 120-second per-call timeout.
• A Limitless Pendant is required to retrieve meaningful data from the API.
• The MCP config must not contain comments; use placeholders like <YOUR_LIMITLESS_API_KEY_HERE> only as references until you substitute your real values.
• The MCP tools support listing and searching recent logs; full-history search may not be available. Use listing tools first to narrow down results before searching.
• Transport is via stdio; ensure your MCP client supports spawning a stdio server and passing environment variables.