krumjahn/applehealth
Privacy-first Apple Health export analyzer for CSV/JSON, charts, and local or cloud AI chat.
What it solves
It converts Apple Health data exports (XML/ZIP) into structured, readable CSV and JSON files. It allows users to bypass the limitations of the raw Apple Health export format by providing a way to analyze, visualize, and query their health metrics locally without needing to upload sensitive data to third-party cloud services.
How it works
The tool uses a streaming converter to process large XML files efficiently, keeping memory usage low. Once converted, users can interact with their data through a command-line chat interface. This interface supports slash commands for specific metrics like heart rate or sleep, and can be paired with over 50 AI models—including local models run via Ollama—to answer questions about health trends in plain English.
Who it’s for
This project is designed for developers, biohackers, and researchers who want full control over their health data pipeline. It is ideal for users who want to perform private, source-aware analysis of their fitness history, reconcile overlapping data from multiple devices, or build custom training plans based on their own biological patterns.
Highlights
- Streaming Conversion: Efficiently turns massive Apple Health exports into usable CSV/JSON files with minimal memory footprint.
- 100% Private Option: Supports fully local AI analysis using models like DeepSeek-R1 or Llama 3 via Ollama, ensuring no data leaves your machine.
- Source Reconciliation: Automatically handles overlapping data from multiple devices (e.g., Apple Watch vs. iPhone) with configurable source priority.
- Interactive Chat: Built-in REPL allows for natural language queries and automated generation of charts for metrics like heart rate, steps, and sleep.
- Extensible: Includes support for OpenClaw skills and provides a clear path for AI agents to programmatically access and analyze your health history.
Related
- Project
- Project
- Project
- Project
- Project