A personal weight and health tracking dashboard for Mac that integrates with Apple Health data, featuring AI coaching, gamification, and activity correlation analysis.
In 2024, I started walking intentionally - tracking every session. After a year, I had 1,000km of data trapped in Apple Health with no good way to analyze it or correlate it with weight changes.
So I built my own dashboard. I'm not a developer by profession - I run enterprise sales for a Fortune 500 company. But I wanted to see if AI tools (specifically Claude) could help a "business person who codes as a hobby" build something genuinely useful.
I didn't write a single line of code. Every component, API route, database migration, and XML parser was generated by Claude Code. I described what I wanted, reviewed the output, tested in the browser, and iterated.
The result: 8,700+ lines of TypeScript across 21 commits - a production-quality health dashboard with features I couldn't find anywhere else. Total time: ~20 hours over 3 weeks.
What surprised me most was the first commit. I described the app I wanted, and Claude Code delivered a working Next.js application with Apple Health XML parsing, SQLite database, interactive charts, and a basic UI - all in one session. The remaining weeks were spent adding features: AI coaching, gamification, nutrition tracking, GPS route visualization, HRV analysis.
This isn't a toy project. It's my actual health dashboard that I use daily.
- Weight Tracking: Import weight data from Apple Health or log manually
- Activity Integration: Steps, workouts, flights climbed from Apple Health
- Body Composition: Body fat % and lean body mass (from smart scales)
- GPS Routes: Walking/running routes visualized on maps
- VO2Max: Cardiovascular fitness tracking
- Resting Heart Rate: Daily resting HR trends
- HRV: Heart rate variability for AI Coach recovery insights
- Personalized Insights: AI Coach powered by Claude analyzes your complete health history
- Recovery Guidance: Uses HRV and resting HR for training recommendations
- Nutrition Sprint: Time-boxed food tracking with AI calorie estimation
- Comprehensive Context: Coach has access to years of data for meaningful insights
- Interactive Charts: Multi-axis charts with 30D/90D/1Y/3Y/5Y/10Y views
- Activity Correlation: Analyzes relationship between activity and weight changes
- Trend Analysis: Milestones, insights, and historical comparisons
- Calendar Heatmaps: GitHub-style activity visualization
Stay motivated with a complete gamification system:
- XP System: Earn experience points for healthy behaviors
- Log weight: +10 XP
- Hit step goal: +15 XP
- Complete workout: +20 XP
- Maintain streak: +5 XP per day
- Levels: Progress through levels as you accumulate XP (Level 1-50+)
- Streaks: Track consecutive days of activity with streak bonuses
- Badges: Unlock achievements for milestones
- First Steps, Week Warrior, Month Master
- Weight milestones (5kg, 10kg lost)
- Consistency badges (30-day, 100-day streaks)
- Daily Quests: Fresh challenges each day to keep you engaged
- Progress Bar: Visual feedback on your journey to the next level
- Energy Levels: Track daily energy 1-10
- Fasting Hours: Log intermittent fasting
- Notes: Add context to any day
Your health data stays on your machine.
- All data stored locally in SQLite database on your computer
- No cloud sync - you control your data completely
- No analytics or telemetry - we don't track usage
- No account required - just clone and run
When you use the AI Coach feature, your health data is sent to Anthropic's Claude API:
| Data Sent | Purpose |
|---|---|
| Weight history | Trend analysis and personalized advice |
| Activity data (steps, workouts) | Activity-weight correlation insights |
| Body composition | Comprehensive health assessment |
| Resting HR, HRV | Recovery and training recommendations |
Important considerations:
- Data is sent only when you explicitly ask the AI Coach a question
- Your API key is stored locally in your database (never transmitted to us)
- Anthropic's data usage policy applies
- You can use the app without the AI Coach feature
- Review Anthropic's privacy policy before using AI Coach
- Don't include sensitive personal information in check-in notes
- Your
.env.localfile (with personal medical context) is gitignored
- Frontend: Next.js 15, React 19, TypeScript, Tailwind CSS, shadcn/ui
- Charts: Recharts
- Database: SQLite with Drizzle ORM
- AI: Anthropic Claude API (Sonnet 4.5)
- Maps: Leaflet for GPS route visualization
- Node.js 18+
- npm or yarn
- Mac (optimized for desktop use)
# Clone the repository
git clone https://github.com/eprouveze/HealthPulse.git
cd HealthPulse
# Install dependencies
npm install
# Start development server (runs on port 4000)
npm run dev- Export your Apple Health data from iPhone (Health app → Profile → Export All Health Data)
- Extract the ZIP file
- Copy the entire
apple_health_exportfolder toimports/in the project - Run the import:
npm run importThe import script:
- Creates automatic backups before importing
- Deduplicates data from multiple devices (iPhone + Apple Watch)
- Imports: weights, steps, workouts, sleep, resting HR, body composition, VO2Max, HRV, GPS routes
- Safe: Settings, goals, and check-ins are never modified
npm run db:generate # Generate migrations from schema changes
npm run db:migrate # Run pending migrations
npm run db:studio # Open Drizzle Studio GUIThe AI Coach requires an Anthropic API key:
- Get your API key from console.anthropic.com
- Click the Settings icon in the app
- Enter your API key and click Save
For personalized AI coaching, create a .env.local file (see .env.local.example):
PERSONAL_MEDICAL_CONTEXT="Your personal health context for the AI coach"This file is gitignored and stays on your local machine only.
src/
├── app/
│ ├── api/ # API routes (weights, stats, coach, etc.)
│ └── page.tsx # Main dashboard
├── components/
│ ├── ui/ # shadcn/ui components
│ ├── weight-activity-chart.tsx
│ ├── gamification-panel.tsx
│ ├── activity-panel.tsx
│ ├── nutrition-sprint.tsx
│ └── lifestyle-tracker.tsx
└── lib/
├── schema.ts # Drizzle database schema
├── db.ts # Database connection
├── analysis.ts # Trend analysis logic
└── gamification.ts # XP and badge calculations
scripts/
├── import-apple-health.ts # Apple Health XML parser
├── generate-demo-data.ts # Generate demo data for testing
└── watch-import.ts # Auto-import watcher
Built by Emmanuel Prouvèze - enterprise sales leader, hobby coder, AI enthusiast.
MIT