Help us improve
Share bugs, ideas, or general feedback.
From cm
Guides code review lifecycle: requesting reviews with pre-checks, receiving feedback with technical rigor, and completing branch integration. Use when finishing tasks or handling PR feedback.
npx claudepluginhub tody-agent/codymaster --plugin cmHow this skill is triggered — by the user, by Claude, or both
Slash command
/cm:cm-code-reviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- **Use when** task complete, before/during PR review
Guides rigorous code review practices: evaluate feedback technically, request code-reviewer subagent reviews after tasks/features, verify before completion claims. For subagent-driven dev and PRs.
Dispatches a code-reviewer subagent after completing tasks or major features to catch issues before merging.
Requests code review via a subagent after completing tasks, implementing features, or before merging to catch issues early. Accepts git SHAs and descriptions for focused evaluation.
Share bugs, ideas, or general feedback.
Full review lifecycle in one skill: Request → Receive → Integrate.
Mandatory:
cm-executionOptional but valuable:
Get git SHAs:
BASE_SHA=$(git rev-parse HEAD~1)
HEAD_SHA=$(git rev-parse HEAD)
Dispatch reviewer subagent with:
Act on feedback:
When receiving feedback — whether from human reviewers, AI reviewers, or code review subagents.
1. READ feedback completely before responding
2. UNDERSTAND the technical reasoning
3. VERIFY if the feedback is technically correct
4. RESPOND with evidence, not agreement
| Feedback Type | Response |
|---|---|
| Technically correct | Fix it. Thank reviewer. |
| Unclear intent | Ask for clarification with specific questions |
| Technically questionable | Challenge with evidence (code, tests, docs) |
| Stylistic preference | Discuss trade-offs, defer to team convention |
❌ "Good catch! Fixed." (without verifying it's actually a problem)
✅ "I verified this: [evidence]. The suggestion is correct because [reason]. Fixed."
✅ "I investigated this: [evidence]. The current code is correct because [reason]."
When implementation is complete and all tests pass.
Verify current state:
npm run test:gate # All tests must pass
git status # Working tree should be clean
Present options to user:
| Option | When | Command |
|---|---|---|
| Merge to main | Feature ready | git checkout main && git merge feature-branch |
| Create PR | Needs team review | git push origin feature-branch |
| Keep working | More tasks remain | Continue on branch |
| Cleanup only | Abandoned/merged | git worktree remove path |
Execute chosen option
Cleanup:
cm-identity-guard before git pushAfter processing review feedback, ALWAYS update .cm/CONTINUITY.md:
[Decision]: [Rationale] — scope: [global|module:{name}]Anti-duplicate: If similar learning exists, reinforce it instead of creating new.
Token savings: Future code reviews in same project avoid repeating the same feedback. Reviewer patterns become accumulated knowledge.
| Skill | Relationship |
|---|---|
cm-execution | Reviews after each task in execution |
cm-quality-gate | Tests must pass before finishing branch |
cm-identity-guard | Before git push |
cm-execution | Cleanup isolated workspace after completion |
Block any PR/diff that violates these regardless of green tests:
Review heuristic: if you can't answer "what task forced this line?" for every changed line, the diff is too wide.
Review early. Verify feedback. Ship with evidence, not hope.