Feedback

Consume the review comments already on a PR — Copilot and human reviewers — and sort them into what's worth fixing and what isn't. This is judgment, not a fresh review: each comment is verified against the current code before a verdict.

Read-only. Never reply to threads, resolve conversations, or post anything. The output is a decision aid for the user.

Distinct from code-review, which generates new findings: this skill evaluates existing reviewer comments.

Workflow

1. Fetch with gh — pull the PR's review comments (default to the current branch's PR). Cover both inline thread comments and top-level review summaries, from Copilot and human reviewers. Keep only unresolved threads — resolution state lives in the GraphQL reviewThreads field (isResolved), not REST pulls/comments, so reach for gh api graphql when you need to filter out resolved/outdated noise. Capture author, path, line, and body for each. No unresolved comments → report that and stop.

2. Verify each against current code — read the file at path:line and confirm the comment still applies. Reviewers (especially bots) comment on stale diffs and miss surrounding context. An outdated: true thread defaults toward Skip unless the concern still holds in the current code. Completion: every comment has been checked against the actual code, not judged from its text alone.

3. Classify each comment using the rubric below. Every verdict carries a one-line rationale grounded in what you saw in step 2.

4. Output the triage table — grouped Address-first (see Output). Completion: every fetched comment appears in the table with a verdict and rationale.

5. Offer to fix — ask whether to implement the Address items. On yes, call the Skill tool with tdd for them (test-first). Don't auto-implement; don't touch Skip/Optional/Discuss items.