Skip to main content
Glama
deenesjakoruzh
aviman1109

livetrack-mcp

by aviman1109
OverviewSchemaRelated ServersScoreDiscussions
Python
Hybrid

livetrack-mcp

自主 MCP 服务器，用于轮询 Garmin LiveTrack，将时间序列存储在 SQLite 中，并通过 claude-runner 每 10 分钟触发一次 Claude 分析。

架构

Garmin LiveTrack URL
    │
    │  (poll every 60 s)
    ▼
livetrack-mcp (port 38100)
    ├── poller.py      — fetch trackpoints from Garmin API
    ├── store.py       — SQLite time-series persistence (/data/livetrack.db)
    ├── tracker.py     — asyncio scheduling + race-end detection
    └── analyzer.py    — build prompt, call claude-runner
         │
         │  POST /run (fire-and-forget)
         ▼
    claude-runner (port 38095)
         │
         │  claude -p <analysis prompt>
         ▼
    Claude (sonnet)
         ├── analyze timeseries
         ├── curl POST /control if thresholds need adjustment
         └── mcp__telegram__send_message → coaching push

核心设计： livetrack-mcp 是完全自主的——比赛期间无需保持 Claude 会话处于活动状态。Claude 作为无状态分析函数每 10 分钟被调用一次。如果 claude-runner 暂时不可用，下一个分析周期会自动重试。

Related MCP server: Garmin-Strava-mcp

MCP 工具

工具

描述

start_tracking(url, race_config)

开始轮询 LiveTrack 分享 URL

stop_tracking()

停止轮询（比赛结束时也会自动停止）

get_tracking_status()

活动状态、已用时间、陈旧时间、轮询错误

get_timeseries(minutes=10)

来自 SQLite 的近期数据

update_thresholds(updates)

在比赛中途更新心率/功率阈值

trigger_analysis()

手动按需分析，绕过计划任务

自定义 HTTP 端点

端点

方法

描述

/control

POST

在比赛中途更新阈值（由 Claude 通过 Bash 工具中的 curl 调用）

/health

GET

健康检查 — 跟踪状态 + 存储统计信息

/control 用法（来自 Claude 的分析提示词）

curl -sf -X POST http://localhost:38100/control \
  -H 'Content-Type: application/json' \
  -d '{"power_max": 150}'

允许的字段：hr_max, hr_min, power_max, power_min, cadence_min, run_hr_max, run_hr_min, run_cadence_min

race_config 字段

字段

类型

默认值

描述

hr_max

int

—

骑行心率上限 (bpm)

hr_min

int

—

骑行心率下限

power_max

int

—

ERG 功率上限 (瓦特)

power_min

int

—

ERG 功率下限

cadence_min

int

—

最低骑行踏频 (rpm)

run_hr_max

int

—

跑步心率上限

run_hr_min

int

—

跑步心率下限

run_cadence_min

int

—

最低跑步步频 (spm)

poll_interval_secs

int

60

轮询 LiveTrack 的频率

analyze_interval_secs

int

600

触发 Claude 分析的频率

analyze_window_min

int

10

传递给 Claude 的数据窗口（分钟）

完整铁人三项的 race_config 示例

{
  "race_name": "CT2026",
  "race_type": "triathlon",
  "hr_max": 144,
  "hr_min": 115,
  "power_max": 165,
  "cadence_min": 82,
  "run_hr_max": 152,
  "run_hr_min": 125,
  "run_cadence_min": 165,
  "poll_interval_secs": 60,
  "analyze_interval_secs": 600,
  "analyze_window_min": 10
}

比赛结束检测

服务器在以下情况自动停止：

  • ≥ 15 分钟没有新的轨迹点 (STALE_STOP_MIN)

  • 且总耗时 ≥ 30 分钟 (MIN_ELAPSED_MIN)

这处理了 Garmin 24 小时 URL 延迟问题：URL 在比赛后仍然有效，但当运动员完成比赛时，新的轨迹点会停止到达。30 分钟的最小值防止了在开始时 GPS 数据稀疏导致的错误停止。

配置（环境变量）

变量

默认值

描述

PORT

38100

服务器端口

HOST

0.0.0.0

绑定地址

MCP_PATH

/mcp

MCP 端点路径

DB_PATH

/data/livetrack.db

SQLite 数据库路径

CLAUDE_RUNNER_URL

http://localhost:38095

claude-runner 基础 URL

RUNNER_WORKSPACE

training

claude-runner 任务的工作区

LOG_LEVEL

INFO

日志级别

OTEL_EXPORTER_OTLP_ENDPOINT

—

OpenTelemetry 收集器 URL（可选）

部署

cd ~/ai-platform/mcps

# Build and start
docker compose up -d --build livetrack-mcp

# Logs
docker compose logs -f livetrack-mcp

# Restart
docker compose restart livetrack-mcp

# Health check
curl http://localhost:38100/health

典型会话（通过训练工作区中的 Claude）

# Start tracking
use_mcp_tool livetrack-mcp start_tracking \
  url="https://livetrack.garmin.com/session/.../token/..." \
  race_config={"hr_max": 144, "power_max": 165, "run_hr_max": 152}

# Check status
use_mcp_tool livetrack-mcp get_tracking_status

# Manual analysis trigger
use_mcp_tool livetrack-mcp trigger_analysis

# Stop (or let it auto-stop)
use_mcp_tool livetrack-mcp stop_tracking

项目结构

livetrack_mcp/
├── Dockerfile
├── pyproject.toml
├── README.md
└── src/livetrack_mcp/
    ├── __init__.py
    ├── __main__.py
    ├── otel.py        # OpenTelemetry setup
    ├── poller.py      # Garmin LiveTrack URL parsing + HTTP fetch
    ├── store.py       # SQLite time-series (sqlite3 + asyncio.to_thread)
    ├── tracker.py     # Scheduling (asyncio.create_task) + race-end detection
    ├── analyzer.py    # Prompt builder + claude-runner caller
    └── server.py      # FastMCP tools + /control + /health

This server cannot be installed

A
license - permissive license
-
quality - not tested
D
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • claude-garmin

    Jack-Abyss
    A
    license
    -
    quality
    B
    maintenance
    MCP server that connects Garmin Connect data to Claude, enabling training analysis, recovery checks, and personalized plans based on real metrics like HRV, training load, and activities.
    12
    MIT

  • DkwtMCP

    bynoe17
    A
    license
    A
    quality
    B
    maintenance
    Local MCP server that connects Claude Desktop with Garmin and Apple Health data to read training and recovery, estimate heart rate and pace zones, analyze performance, and create structured workouts.
    22
    MIT

View all related MCP servers

Related MCP Connectors

  • Garmin — MissingMCP

    Garmin data in Claude: 135 tools — activities, sleep, HRV, training, workouts. Free, open source.

  • Fitness AI Connector

    Garmin data in Claude & ChatGPT via the Garmin Health API. OAuth sign-in, no password sharing.

  • Withings Health Data

    MCP server for Withings health data — sleep, activity, heart, and body metrics.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/aviman1109/livetrack_mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server