A training copilot, gone AI-native. feezify reads your day — your objective form (training load) crossed with how you actually feel — and tells you, in plain words, whether today is green, amber, or red, and why. It does not prescribe workouts. You decide; it helps you read yourself more clearly.

It installs on your own AI as a skill, reads a folder of markdown you own, and runs entirely on your machine — no server, no account, no telemetry.

🇫🇷 Version française : fr/README.md


Start here

Pick the door that fits you — you don’t need to read everything.

If you… Read
just want to know what this is What is feezify?
don’t want to touch a terminal Install — no terminal (3 steps)
are a developer / OpenClaw user Install — developer path
are set up and want to use it Daily use
want to understand how it reads you The method
care about your data Privacy & consent
are stuck or have a question FAQ & troubleshooting

The whole thing in one minute

  1. Connect Strava (optional) so feezify can see your training load — via Claude’s official Strava connector (no keys), or your own tokens on the developer path.
  2. Install the skill on your AI (upload a zip in Claude, or point a dev setup at the SKILL.md).
  3. Keep a short journal — a few honest lines a day (sleep, fatigue, mood, a niggle).
  4. Ask “how am I today?” — feezify runs a deterministic read and answers green / amber / red and why. It never prescribes.

Everything below just expands those four steps.

The documents

Canonical sources in the repo

These docs are the reader’s manual. The source of truth for each subject lives with the code and wins in case of conflict:

Status & license

Early build — beta, feedback welcome. Open an issue with what reads true and what doesn’t. Licensed Apache-2.0. Built by Nicolas Jouanno — see author.md.