aitime-calc

Calculateur de temps humain équivalent écoulé depuis la sortie d'un modèle/outil IA — accélération perçue du progrès IA vs cycle de maturation classique (ex: packages npm).

Exemple: Cursor YOLO mode sorti en 2024 → ressenti comme ~27 ans de temps humain aujourd'hui.

Fleet: founder (console) → cto (lead) → dev × backend/frontend, orchestrée via niwa + agent-relay.

Status

En cours de conception — le CTO de la fleet fait la recherche sur le concept avant implémentation. Chaque lead commit sous son propre nom.

API reference

See docs/API.md for the JSON endpoints.

Development / gate parity

Run bash scripts/gate-check.sh before qa-submit — it mirrors the reviewer's clean checkout (fresh npm ci for api/ and web/, then typecheck, test, build) so native-dep / missing-asset failures surface locally instead of at the gate.

Gate smoke test

This line was added to exercise the QA gate end to end: PR creation, the reviewer verdict mirrored onto the PR, and the merge gated on GitHub checks.

PR gate demo (App path, clean body)

This change exists to watch the gate do the whole chain by itself: open a real PR with a readable body, review it, gate the merge on checks.