The Garmin MCP server offers extensive coverage of Garmin Connect data, ideal for users wanting to integrate fitness and health data into AI workflows. Installation is straightforward, supports multiple clients, but requires attention to authentication and HTTP security.
Strongest · Documentation 17/20 Weakest · Reliability 11/20
Static review · not runListed 2026-08-07
Read the FMRS scoring method →Runtime：Python 3.12+
GARMIN_EMAIL
Garmin Connect login email; use GARMIN_EMAIL_FILE instead in Docker-style setups, the two are mutually exclusive.
GARMIN_EMAIL_FILE
Path to a file containing the Garmin Connect email.
GARMIN_PASSWORD
Garmin Connect password; use GARMIN_PASSWORD_FILE instead in file-based secret setups, the two are mutually exclusive.
GARMIN_PASSWORD_FILE
Path to a file containing the Garmin Connect password.
GARMIN_IS_CN
Set to true to use Garmin Connect China (garmin.cn); defaults to false.
GARMIN_FIT_DOWNLOAD_DIR
Default directory for downloaded activity files; skips the first-run setup prompt in download_activity_file.
GARMIN_FIT_CONFIG
Path to the persisted download-directory config file (default ~/.garminconnect_fit_config.).
GARMIN_ENABLED_TOOLS
Comma-separated tool allowlist — only these tools register.
GARMIN_DISABLED_TOOLS
Comma-separated tool denylist — these tools are skipped; ignored when an allowlist is set.
GARMIN_MCP_TRANSPORT
Transport mode: stdio (default), streamable-http, or sse.
GARMIN_MCP_HOST
Bind address for HTTP transports, default 127.0.0.1.
GARMIN_MCP_PORT
Bind port for HTTP transports, default 8000.
garmin-mcp-auth to save OAuth tokens to ~/.garminconnect.uvx command.{
"mcpServers": {
"garmin": {
"command": "uvx",
"args": [
"--python",
"3.12",
"--from",
"git+https://github.com/Taxuspt/garmin_mcp",
"garmin-mcp"
]
}
}
}
Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.
{
"servers": {
"garmin": {
"command": "uvx",
"args": [
"--python",
"3.12",
"--from",
"git+https://github.com/Taxuspt/garmin_mcp",
"garmin-mcp"
]
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
claude mcp add garmin -- uvx --python 3.12 --from git+https://github.com/Taxuspt/garmin_mcp garmin-mcp
Run it in a terminal; replace any <…> placeholders with your own values first.
Run garmin-mcp-auth once to save OAuth tokens, restart the client, and confirm the garmin server's tools appear in the tool list; then ask 'How many steps did I take yesterday?' — a data-backed answer confirms the connection.
Once connected, you can ask your AI assistant things like:
Listed from the project's documentation, not tested by this site.
The Garmin MCP server connects to Garmin Connect and exposes your fitness and health data to Claude and other MCP-compatible clients. It leverages the python-garminconnect library and offers 110+ tools covering activity management, health metrics, training load trends, power analysis, and more, including advanced features like power duration curve, FIT file parsing, and DI2 electronic shift intelligence. It supports tool filtering via environment variables and provides high-level workout builder tools.
Manage your Hevy workout data from AI assistants via MCP.
Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
One binary. One grammar. Evidence from the biomedical sources you already trust.
Encrypted, fully offline long-term memory for AI agents
Source revision cfc5d799ab0f Data synced 2026-10-12 Read the FMRS scoring method