failure-flywheel
Installation
SKILL.md
Skill: failure-flywheel
Turn recurring failures into rules. Capture by code, count deterministically, draft a stub for a human to approve — never auto-promote.
When to use
- A task failed in a way that could recur (invented an API, broke a test, ignored a rule, over-built). Record it so it is counted, not forgotten.
- Periodically (end of session, or every N failures):
--reportto see which failure class recurs most. - When a class crosses the recurrence threshold:
--draft <category>to seed a candidate rule, then run/proposeto finish it.
Do NOT use to auto-write rules. This skill seeds the gate; a human always approves.
The flywheel (Hamel: create -> label -> fix -> repeat)
The slow human step is "notice the same mistake keeps happening and turn it into a rule." This makes the mechanical 90% deterministic (capture + taxonomy + count + scaffold) and refuses to fake the 10% that needs judgement (what the rule actually is).
Commands
# 1. CAPTURE — append one failure (BY CODE; failures.jsonl is gitignored local history)
python3 harness/scripts/failure-flywheel.py record <category> "<summary>" [--detail "..."]