01
What this skill does
Use this skill when a change set needs a user-visible behavior gate. It reviews a declared scope, separates intended visible changes from regressions, and writes a report that maps every touched surface to reviewed, intentional, not covered, or not relevant.
02
When to use it
- SignalChecking whether refactors or feature work broke user-facing flows.
- SignalAuditing loading, error, permission, retry, ordering, export, email, or CLI-output changes.
- SignalProducing a review artifact with severity tied to the strongest unresolved finding.
03
How it works
- 01
Set or infer the review scope and read requirements when available.
- 02
Map touched user-visible surfaces before judging behavior.
- 03
Compare current behavior against baseline, intent, and user expectations.
- 04
Write all distinct findings, not only the top few.
- 05
Record intentional visible changes and uncovered surfaces in the coverage ledger.
04
What you get back
- SignalA Markdown regression-review report.
- SignalA gate recommendation aligned to findings and coverage.
- SignalA complete findings index and coverage ledger.
05
Important boundaries
- SignalDo not silently sample a large review scope.
- SignalDo not treat implementation ugliness as a regression unless it changes visible behavior.
- SignalUse `hack-review` instead when the main concern is brittle implementation structure.
06
Files worth opening
Workflowskills/regression-review/SKILL.mdScope, user-visible findings, coverage, and gate rules.Report templateskills/regression-review/references/report-template.mdCanonical report sections and coverage ledger shape.Runtime metadataskills/regression-review/agents/openai.yamlOptional agent runtime metadata for this skill.