| name | quality-assurance |
| description | Reviewing whether a change carries adequate verification evidence — "is this verified", "did this break anything", "were the required checks run". The reviewer's QA pass on top of the development verification rules. Covers requiring command evidence for the format and lint gate, treating a change to the gate's own configuration as a risk to the gate itself, matching manual checks to the changed output surfaces, mapping skipped checks to residual risk, demanding second-pass verification after severe findings, and asking whether a check that passed was ever capable of failing. |
| user-invocable | false |
Quality Assurance
Use this capability to judge whether a change has been adequately verified before merge. This is the reviewer's lens — flag missing evidence and link to the developer-facing rule rather than re-deriving it.
The severity labels used throughout (Critical, Major, Minor) are owned by the project's review severity model; consult it for each tier's definition, fixed floors, and verdict mapping.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119.
Verification Evidence
See verification-evidence.md for:
- Commands run, exit status, and relevant output
- The evidence-adequacy decision flow from changed surface to covered-or-flag
- Whether a result was ever capable of coming out differently, and the fixture and harness failures that make a green or red check carry no information
- Manual checks matched to changed output surfaces
- Skipped required checks and residual risk
- Second-pass verification after fixing Critical or Major findings
Lint and Format Gate
See lint-and-format-gate.md for:
- The author ran the format and lint commands per the project's code-quality rules
- A diff touching the gate's own configuration, hooks, or CI workflows, and the evidence that the gate still catches violations
- No new inline linter suppressions without an inline justification
- No new lint warnings introduced into modified files
Manual Verification Evidence
See manual-verification.md for:
- The author exercised non-default content states when the change touches a data-driven surface
- The not-found UI was verified for routing changes
- The dev-server output was checked for new warnings or errors
