garmin-mcp
Privacy-focused Garmin Connect MCP server for triathlon training, enabling analysis of health metrics, activities, training load, and structured workout creation.
README
garmin-mcp
A privacy-focused Garmin Connect MCP server for triathlon training. ~47 tools covering daily health metrics, activity analysis, training load, running, cycling, swimming, strength training, and workout creation — plus triathlon-specific analysis (brick workouts, sport volume balance, cross-sport fitness snapshots).
Deployed as a remote MCP server over HTTPS. Connects to Claude Desktop or Claude Code via the streamable-HTTP transport.
One-Time Auth Setup
Run this locally once to generate tokens:
-
Install locally:
pip install -e . -
Authenticate with Garmin:
garmin-mcp-authSaves tokens to
~/.garminconnect/garmin_tokens.jsonand prints the base64 export command. -
Copy the printed
GARMIN_TOKENS_JSON=...value — you'll need it for the secrets step below.
Deploy to Fly.io
Prerequisites: flyctl installed and authenticated (flyctl auth login).
-
Register the app name (first time only):
flyctl apps create garmin-mcpIf
garmin-mcpis taken on Fly.io (global namespace), use a unique name likegarmin-mcp-yournameand update theappfield infly.tomlto match. -
Set secrets:
flyctl secrets set \ GARMIN_EMAIL=you@example.com \ GARMIN_TOKENS_JSON=<base64-from-auth-step> \ MCP_AUTH_TOKEN=$(openssl rand -hex 32)Save the generated
MCP_AUTH_TOKENvalue — you'll need it for the Claude Desktop config. -
Deploy:
flyctl deploy -
Verify:
curl https://garmlink.fly.dev/healthExpected:
{"status":"ok"}
Auto-Deploy via GitHub Actions
After the first manual deploy, subsequent pushes to main auto-deploy via GitHub Actions.
Add FLY_API_TOKEN to your GitHub repo secrets:
flyctl tokens create deploy
Copy the output, then go to: GitHub repo → Settings → Secrets and variables → Actions → New repository secret → name it FLY_API_TOKEN.
Claude Desktop Config
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"garmin": {
"url": "https://garmlink.fly.dev/mcp",
"headers": { "Authorization": "Bearer <MCP_AUTH_TOKEN>" }
}
}
}
Replace <MCP_AUTH_TOKEN> with the value you set in the secrets step.
Coaching Skills
Run these slash commands from this project directory in Claude Code:
| Command | Purpose |
|---|---|
/morning-check |
Daily readiness briefing (HRV, sleep, body battery) |
/analyze-week |
Weekly training load and sport balance review |
/race-readiness |
Pre-race fitness assessment across all disciplines |
/create-workout |
Guided structured workout builder → pushes to Garmin |
Environment Variables
| Variable | Description |
|---|---|
GARMIN_EMAIL |
Your Garmin Connect email |
GARMIN_TOKENS_JSON |
Base64-encoded token file (from garmin-mcp-auth) |
MCP_AUTH_TOKEN |
Required. Bearer token protecting the MCP endpoint; must be at least 32 characters. The server refuses to start without it. |
GARMIN_PASSWORD |
Optional. Only used to re-authenticate if the stored tokens expire. |
ALLOW_UNAUTHENTICATED |
Set to 1 to run with no authentication. Localhost development only — never on a public address. |
PORT |
Server port (default: 8000, set automatically by Fly.io) |
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。