Install Garmin Connect MCP

Pick your client, copy the command, done.

Manual setup required. The maintainer's config contains paths only you know — edit the placeholders below before adding it to Claude Code.

Run this once before adding it to Claude Code.

git clone https://github.com/leewnsdud/garmin-connect-mcp.git
cd garmin-connect-mcp
uv sync

claude mcp add garmin-connect-mcp -- "/Users/<username>/.local/bin/uv" --directory /path/to/garmin-mcp run garmin-mcp

Replace any placeholder paths in the command with the real path on your machine.

Environment Variables

Set these before running Garmin Connect MCP.

VariableDescriptionRequired

GARMIN_EMAILGarmin Connect email addressNo

GARMIN_PASSWORDGarmin Connect passwordNo

GARMIN_TOKEN_DIRDirectory path to store OAuth tokensNo

Once configured, Garmin Connect MCP gives your AI agent access to:

get_recent_activitiesRetrieve recent running activities including performance metrics.

count

get_activity_detailGet detailed analysis for a specific activity.

activity_id

get_training_statusCheck current training status.

date

create_running_workoutCreate and upload a structured workout to Garmin Connect.

namestepsdescription

get_daily_wellnessRetrieve stress, body battery, SpO2, and respiratory data.

date

After setup, try these prompts with your AI agent:

→Analyze my last 5 runs and tell me if my training load is increasing too quickly.

→What is my current VO2max and how does it compare to my race predictions?

→Create a structured workout for tomorrow: 10 minutes warmup, 4x1km intervals at 4:30 pace, and 5 minutes cooldown.

→Summarize my body battery and stress levels for the past week to see if I am recovering well.

→Show me my personal records for 5K and 10K distances.

Save the docs, env vars, and workflow around Garmin Connect MCP in Conare so Claude Code, Codex, and Cursor remember it next time.