# Taxuspt/garmin_mcp

MCP server to access Garmin data

Repository: https://github.com/Taxuspt/garmin_mcp
Canonical: https://ross.abutalabs.com/products/garmin_mcp
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-04T18:29:56+00:00

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 96, release rhythm 51, longevity 38
- inputs: {"age_days": 543, "days_push": 29, "days_rel": 117, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1065, forks 335 (observed 2026-08-28T04:03:26.518808+00:00)

## What it is
A Model Context Protocol (MCP) server that connects to Garmin Connect and exposes fitness and health data (activities, sleep, heart rate, training metrics) to Claude and other MCP-compatible clients. It implements 110+ tools covering roughly 90% of the python-garminconnect library, including read and write access to activities, workouts, gear, and nutrition.

## Use cases
- ask Claude about my recent Garmin activities
- analyze my sleep and heart rate data from Garmin
- track my training load and VO2 max trends
- edit activity names and notes via AI assistant
- download my Garmin activities as GPX or FIT files
- create and schedule workouts without writing JSON
- review my weekly steps and stress summary

## When to choose
- you use Claude or another MCP-compatible client and want your Garmin fitness data accessible in conversations
- you want AI-assisted analysis of training metrics like CTL/ATL/TSB, HRV, and power curves
- you need to manage Garmin activities, workouts, or gear programmatically through natural language

## When to avoid
- you need direct programmatic API access without an LLM client - use python-garminconnect directly
- you don't have a Garmin Connect account or compatible device
- you need real-time streaming of live workout data

## Facets
- artifact type: service
- maturity: active
- function: mcp, sdk, analytics, data-science
- domain: sports, healthcare, large-language-models
- platform: python, self-hosted, cross-platform
- tags: garmin-connect, mcp-server, fitness-data, health-metrics, claude, wearables, training-analytics, ai-agents

## Member repositories
- Taxuspt/garmin_mcp (main) score 69

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:26.518808+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:55:28.259621+00:00, confidence not recorded.
  - readme: https://github.com/Taxuspt/garmin_mcp (fetched 2026-08-28T04:03:26.518808+00:00, sha b7d4efaabc86)
- Data as of 2026-08-30T08:39:29.467469+00:00.