mcpField
A
apple-health-skill

apple-health-skill

github.com/nftechieOther343 installsUpdated 37w ago
Repository

About apple-health-skill

A skill for querying Apple Health data through Transition using an AI agent. It supports natural-language questions about workouts and health metrics, alongside workout, performance, and profile endpoints.

Key features

  1. 01AI chat grounded in Apple Health data
  2. 02Workout retrieval by date range
  3. 03CTL, ATL, and TSB performance chart
  4. 04Performance statistics and athlete profile
  5. 05Unauthenticated workout-of-the-day endpoint

Use cases

  1. 01Asking about heart-rate or sleep trends
  2. 02Comparing workout volume or running pace
  3. 03Checking fitness, fatigue, and form metrics
  4. 04Generating a quick structured workout

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/nftechie/apple-health-skill --skill apple-health-skill

After installing, type/apple-health-skillin Claude Code to confirm it loaded.

Bundled filesREADME.md

Details

From the project README

Apple Health Skill

Chat with your Apple Health data using AI. Ask about your workouts, heart rate trends, activity rings, VO2 Max, and more. Powered by Transition, which syncs with Apple Health to give AI agents access to your fitness data.

Setup

  1. Download Transition and grant Apple Health access
  2. Go to Settings > API Keys and tap Generate New Key
  3. Set the environment variable:
export TRANSITION_API_KEY="tr_live_xxxxxxxxxxxxxxxxxxxxx"

No Auth Required

Workout of the Day

Generate a random structured workout — no account needed.

curl "https://api.transition.fun/api/v1/wod?sport=run&duration=45"

Parameters:

  • sport — run, bike, swim, or strength (default: run)
  • duration — minutes, 10-300 (default: 45)

Authenticated Endpoints

Base URL: https://api.transition.fun Auth: Pass X-API-Key header on every request.

AI Coach Chat

Ask questions about your Apple Health data. The AI coach has full context on your workouts and health metrics.

curl -X POST -H "X-API-Key: $TRANSITION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"message": "How has my resting heart rate changed over the last month?"}' \
  "https://api.transition.fun/api/v1/coach/chat"

Example questions:

  • "How many workouts did I do this week?"
  • "What's my VO2 Max trend?"
  • "How has my sleep been trending this week?"
  • "Compare my running pace this month vs last month"
  • "Should I take a rest day based on my recent training?"
Get Workouts

Retrieve scheduled workouts for a date range.

curl -H "X-API-Key: $TRANSITION_API_KEY" \
  "https://api.transition.fun/api/v1/workouts?start=2026-02-09&end=2026-02-15"

Parameters:

  • start — Start date (YYYY-MM-DD, required)
  • end — End date (YYYY-MM-DD, required)
  • Maximum range between start and end is 90 days.
Performance Management Chart (PMC)

Get CTL (fitness), ATL (fatigue), and TSB (form) calculated from your Apple Health workouts.

curl -H "X-API-Key: $TRANSITION_API_KEY" \
  "https://api.transition.fun/api/v1/performance/pmc"
Performance Stats

Get FTP, threshold paces, heart rate zones, and other metrics derived from your Apple Health data.

curl -H "X-API-Key: $TRANSITION_API_KEY" \
  "https://api.transition.fun/api/v1/performance/stats"
Athlete Profile
curl -H "X-API-Key: $TRANSITION_API_KEY" \
  "https://api.transition.fun/api/v1/profile"
Chat History
curl -H "X-API-Key: $TRANSITION_API_KEY" \
  "https://api.transition.fun/api/v1/coach/history"

Rate Limits

TierRead EndpointsAI Endpoints
Free100/day3/day
Paid10,000/day100/day

Tips for Agents

  1. Use coach chat as the primary interface. It has full context on the user's Apple Health workouts, heart rate, and performance — just ask natural questions.

  2. Check fatigue before recommending hard workouts. Call GET /api/v1/performance/pmc and look at TSB. If TSB is below -20, the athlete is fatigued.

  3. Use the free WOD endpoint for quick workouts. No auth needed — great for users who just want a workout suggestion.

  4. Date format is always YYYY-MM-DD for all date parameters.

Comments

Markdown supported