Combines normalized HealthKit-style activity, nutrition, Strong workout data, and body scans to explain whether fat loss occurred while strength was preserved. I used Codex with GPT-5.6 to implement normalization, EWMA, estimated 1RM, privacy checks, tests, and the product experience. A future live path can use GPT-5.6 to write the brief after local aggregation; the public demo uses synthetic tested data, and raw health data remains local.

Built With

Share this project:

Updates