bmad-code-review

Trigger the workflow by saying "run code review" and specify the target: uncommitted changes, staged changes, a branch diff against a base branch, a commit range, or a pasted diff. The skill constructs the diff with git and launches parallel review layers.

It uses three parallel layers: a Blind Hunter that sees only the diff, an Edge Case Hunter with project read access, and an Acceptance Auditor that checks the diff against a spec file and context docs when provided.

Yes. The workflow supports a no-spec mode where the Acceptance Auditor layer is skipped. Findings that would require a decision are reclassified as patch or defer since there is no spec to resolve ambiguity against.

Findings are normalized, deduplicated, and sorted into four buckets: decision-needed (requires human input), patch (unambiguous fix), defer (pre-existing issue), and dismiss (noise). Dismissed findings are dropped, and patches can be applied automatically.

Yes. When a story file and sprint-status.yaml exist, the workflow writes findings into the story file, updates the story status to done or in-progress based on resolution, and syncs the corresponding entry in sprint-status.yaml.

The failed layer name is recorded and the review proceeds with findings from the remaining layers. If zero findings remain and any layer failed, the user is warned that the review may be incomplete rather than reporting a clean pass.