Debug incident
Finish with Proof: a portable execution protocol for coding agents—diagnose before patching, bound delegation, and verify real outcomes.
npx -y skills add MiaoY0uShan/FP --skill debug-incidentAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Debug-first diagnosis protocol for unknown failures. Use when the cause is unknown, a bug needs diagnosis before fixing, or the user asks 'what's wrong' without requesting a fix. Provides structured hypothesis-probe-decision loop with evidence tracking.
SKILL.md
1.2 KB, as published. Nobody here has run it
FP Debug-First Diagnosis
Load ../templates/debug-incident-checklist.md for the full checklist.
Route
pin symptom → read-only baseline → falsifiable hypothesis
→ cheapest discriminating probe → decision → authorized fix
→ original reproduction + regression + negative control
Rules
- Read-only until a cause is supported and user authorizes a fix.
- Keep at most two active hypotheses; run one discriminating probe at a time.
- Speculative patches are never probes.
- Three consecutive non-narrowing probes → architecture/observability checkpoint.
- After hypothesis is supported, extra probes must change a named decision or fill a named acceptance row. Otherwise stop and reuse evidence.
Active incident variant: OBSERVE → CONTAIN → RESTORE → REPAIR → LEARN
Preserve evidence, access, and rollback. Restore service before permanent repair.