Fitness AI Connector MCP Server

Garmin data in Claude & ChatGPT via the Garmin Health API. OAuth sign-in, no password sharing. Exposed over MCP by the fitness ai connector mcp server, that capability becomes something an assistant can invoke while it works, not something you go and do afterwards.

What it actually does

Connect your Garmin health and fitness data to Claude and ChatGPT through the official Garmin Health API. Hosted remote MCP server (Streamable HTTP) with OAuth 2.0 sign-in — your Garmin password is never shared. 12 tools: daily health summary (sleep, HRV, stress, Body Battery), activities with lap and time-series analysis, body composition, menstrual cycle tracking, trends, and on-demand data refresh. 3,000+ users and 150+ paid subscribers as of July 2026. Free tier available.

Its toolset

Everything the assistant can do here goes through one of these:

• get_user_profile — Garmin connection status, subscription tier, VO2max, and fitness age
• get_health_summary — Daily snapshot: steps, calories, heart rate, sleep stages & score, stress / Body Battery, HRV. Optional opt-in time series (heart rate, stress, Body
• get_activities — List activities (runs, rides, swims) with pace, distance, calories, and heart rate per activity
• get_activity_details — Lap splits, HR drift, pace stability, negative split ratio, cadence and power stats for one activity, plus optional per-second time series
• get_body_composition — Weight, BMI, body fat %, muscle mass, bone mass, body water % (smart scale required)
• get_cycle_data — Menstrual cycle snapshots: phase, day in cycle, period start, cycle length (predicted vs actual), fertile window, with pregnancy semantics when
• get_trends — Weekly averages, directional changes, and pattern detection ("am I improving?")
• refresh_data — Re-fetch a range of Garmin data; dates before your connection date are skipped
• refetch_activity_detail — Request recovery of a missing activity detail
• manage_subscription — Billing portal and account deletion (30-day recovery)
• get_help — Setup, troubleshooting, and billing FAQ answers
• acknowledge_garmin_source — Garmin brand attribution shown when data is first presented

Adding it to your client

Being a remote server, there is no local install. You register the endpoint with your client, authorise it once, and the tools appear.

When to reach for it

Among the payments and commerce options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Fitness AI Connector's toolset — get_user_profile, get_health_summary, get_activities and 9 more — is a fair guide to whether it matches your workflow. It is maintained by fmp-projects; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against Fitness AI Connector's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Caveats

• Your data travels to the provider's service, so the usual questions apply about what you send and what they retain.
• With 12 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Fitness AI Connector.
• MCP clients confirm each tool call by default. Leave that on until you have watched what the fitness ai connector mcp server does with a few real requests.