codexkit-bug-hunt
Reproduce, isolate, patch, and verify defects with explicit expected-versus-observed behavior.
SKILL.mdskills/codexkit-bug-hunt/SKILL.md
GitHubBug Hunt
Use this skill when a report says something is broken and you need disciplined debugging rather than guesswork.
Workflow
- Capture exact observed behavior and expected behavior.
- Reproduce with the cheapest reliable path.
- Narrow the failure boundary before editing code.
- Patch the smallest credible cause.
- Verify the original failure and nearby regressions.
Required outputs
- reproduction path
- root-cause hypothesis
- patch summary
- verification notes
Avoid
- fixing symptoms without a reproduction
- stacking unrelated cleanup into the bug fix
- claiming a root cause without evidence
Quality Criteria
- Every finding is tied to a specific evidence source (log, test, metric)
- Pass/fail criteria are binary and measurable — no subjective judgments
- Severity levels are assigned with clear thresholds
- Remediation steps are provided for all critical and high findings
Verification (4C)
| Check | Question |
|---|---|
| Correctness | Are all pass/fail criteria applied against the correct standard or rule? |
| Completeness | Were all required dimensions or checklist items evaluated? |
| Context-fit | Does the verification scope match the actual risk level of the deliverable? |
| Consequence | If this passed verification but had a hidden flaw, what is the worst-case impact? |
Edge Cases
- Incomplete data for full assessment — Document which checks were limited and flag for re-verification when data becomes available.
- Ambiguous pass/fail criteria — Request clarification from the standard owner before scoring. Mark as 'Needs Review'.
- Multiple overlapping standards — Identify the governing standard and note where others diverge.
Changelog
- v1.0.0 — Initial release