Hey everyone! I built Open Health because I was frustrated that my Apple Watch and iPhone had been collecting millions of data points about my sleep, heart rate, HRV, and workouts — but I couldn't actually use any of it with an LLM. Anthropic and OpenAI both have Apple Health connectors now, but they're US-only. If you're in NZ, the UK, or the EU, you're locked out. And even if you are in the US, you don't control what the model sees or how the data is framed. Open Health gives you that control. Export your Apple Health data (six taps in the Health app), drop the zip in the browser or run the CLI, and you get seven clean markdown files covering your health profile, weekly summaries, workouts, body composition, sleep/recovery, cardio fitness, and a ready-to-paste system prompt. The privacy story is simple: nothing leaves your device. The entire parse runs in a Web Worker in your browser (or locally via the CLI). There's no upload endpoint because there's no server. Open DevTools and watch — the network panel stays empty. If you want to go fully local, pipe the output into Ollama or any local model. Your health data never touches a third-party server. MIT licensed, source on GitHub. Would love to hear what you think!

Forum Threads

p/open-health-unlock-your-apple-health

•

7mo ago

Your Apple Health data is sitting on millions of data points — what would you ask it?

We're launching Open Health soon an open-source tool that turns your Apple Health export into clean markdown files you can feed into any LLM.

Your iPhone and Apple Watch have been quietly collecting data on your sleep stages, heart rate, HRV, workouts, nutrition, and more. But that data is trapped in a massive XML file nothing will open.

View all

Open Health — Unlock Your Apple Health

Chat with your Apple Health data using any LLM

Chat with your Apple Health data using any LLM

Forum Threads

Your Apple Health data is sitting on millions of data points — what would you ask it?

Forum Threads

Your Apple Health data is sitting on millions of data points — what would you ask it?