Failure Drills And Recovery
Named failure modes with rehearsal steps, early signs, guards, and rollback moves.
How drills run
Each drill injects one fault on a branch, watches the guards fire, then recovers to the last kept mainline. Drills run on a schedule so recovery stays muscle memory rather than theory.
Every drill writes the same record shape as a real cycle, with task, trace, score, lesson, patch, and verdict files, so reviewers can audit the rehearsal like production history.
Drill book
Self confirming judge drill feeds weak patches to a judge that approves everything, then checks that calibration sets flag the judge as broken before any promotion. Diversity collapse drill fills the archive with near copies and checks that novelty scoring plus parent sampling restore spread.
Metric gaming drill raises a proxy score while hidden transfer stalls, then checks that the held out gate blocks promotion. Direction drift drill edits goals without review and checks that schedule logs plus human checkpoints reject the run. Frozen path drill proposes a judge edit and checks that the guard rejects by rule with a stored reason.
Drill cards
| Drill | Early sign | Recovery move |
|---|---|---|
| Self confirming judge | Approval rate near total | Recalibrate judge, replay calibration sets |
| Diversity collapse | Archive fills with near copies | Novelty scoring plus fresh parents |
| Metric gaming | Proxy rises, transfer stalls | Block on held out gate, step difficulty down |
| Direction drift | Goals shift without review | Halt schedule, restore last kept mainline |
| Frozen path touch | Patch lists judge or gate files | Reject by rule, store reason |
Recovery commands
git status --short --branch git log --oneline -3 py engine/rsi_loop.py --state engine_state --generation 3 --files worker/planner.py --changes 1 --passed true --gain 0.1 --cost 0.0