Torc is a local-first toolkit for reverse engineering health wearables and building weird/fun things on top of the raw data.
Right now the first target is the Oura Ring 4, because that is what I have. The bigger idea is not "open source Oura", and not just rings either.
The point is to build a clean core for:
- protocol work
- raw-data access
- sync / decode tooling
- my own ML models and derived metrics
- strange and fun interfaces on top
- understand how the device actually talks
- pull data locally and work from first principles
- make my own metrics instead of stopping at the vendor app
- build playful interfaces, not just another dashboard
- keep the code small, legible, and fast to iterate on
- docs/charter.md: for AI agents
- docs/protocol.md: current protocol reference
- docs/findings.md: short confirmed findings and open questions
- Prior community reverse engineering work, especially ringverse, helped establish a useful baseline.