Your WHOOP band doesn't stop working when the subscription does — the app just goes dark. OpenStrap is what I built so a 4.0 band that would otherwise sit in a drawer keeps doing something: read your own data off it over Bluetooth, keep the raw bytes, and turn them into the numbers you actually look at.

Everything runs on your phone. The band talks to the app over Bluetooth, the app decodes the bytes, computes the metrics, and stores the results locally. There's no account, no cloud, and no server that ever sees your raw data. That isn't a privacy setting you switch on — it's the architecture.

Is it a replacement for WHOOP? No, and I'm not going to pretend it is. They've got years of research and a whole team; this is one person and textbook methods. But it's a real second life for hardware you already own, and your data stays yours.

Tested on WHOOP 4.0 only. Once you start using it, don't reconnect the band to the official WHOOP app — a firmware update could change or break the events this relies on.

• iOS — join the TestFlight beta →
• Android — download the APK →

• edge — the phone app, and where everything actually happens: Bluetooth, local storage, the compute pipeline, every screen.
• protocol — the decoders. Raw WHOOP 4.0 record bytes into named fields. Pure Dart, zero dependencies, runs on-device.
• analytics — the math. Published, peer-reviewed methods (Banister TRIMP, Cole-Kripke, Lomb-Scargle and friends) turn heart rate and motion into metrics — each carrying its own confidence and tier, and returning nothing rather than guessing when the data isn't there.
• research — the lab notebook. The protocol written down, plus a one-file Python client so you can talk to a band from a terminal.
• backend — optional, and not required. A small self-hostable companion for importing an old cloud account and for opt-in telemetry. The app doesn't need it and doesn't use it by default.

There are bugs. Please open issues as you find them and I'll work through them. A lot of the protocol — the events especially — is empirical guesswork, and no one person can confirm it to 100%. The more people poke at it, the closer we get to nailing every field down with real confidence.

The metrics are approximations built from published research, not medical-grade measurements. Nothing here is a diagnosis.

Mostly by wearing it on a real wrist and reporting what came out wrong — which is genuinely the most useful thing anyone can do here, since there's one person's physiology in the test data otherwise.

Not affiliated with, endorsed by, or connected to WHOOP. "WHOOP" is their trademark, used only to say which device this talks to. MIT licensed.