An MCP server that synthesizes Garmin Connect training, sleep, and recovery data into compact, verdict-first reports for an LLM. It also lets you browse activities, inspect individual sessions, store athlete profile notes, log RPE/wellness, and trigger syncs or access bounded raw data.
From the registry: A coach's morning report from your Garmin data, for any LLM via MCP.
Please install the `fartlek` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install uvx or pipx** — Install a Python app runner. The README recommends uvx; pipx is also supported. (https://docs.astral.sh/uv/)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["fartlek-mcp"]
Note: One-time authentication is required before normal use: run `fartlek auth` (or `uv run fartlek auth` if running from source) to log into Garmin Connect; tokens are stored locally. Optional env vars mentioned by the README: `GARMINTOKENS` overrides the token location and `FARTLEK_HOME` overrides the data directory, but neither is required for the default setup. The server can also be invoked directly as `fartlek-mcp` after installation.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.GARMINTOKENSOverrides the local token storage location.FARTLEK_HOMEOverrides the local data directory (default ~/.fartlek).AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Persistent codebase knowledge graph. Survives session restarts and context compaction.
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.