Skip to content

๐Ÿ“‹ Full Skill Source โ€” This is the complete, unedited SKILL.md file. Nothing is hidden or summarized.

โ† Back to Skills Library

Code Review โ€” Request + Receive + Complete โ€‹

Full review lifecycle in one skill: Request โ†’ Receive โ†’ Integrate.

Part A: Requesting Code Review โ€‹

When to Request โ€‹

Mandatory:

  • After each task in cm-execution
  • After completing major features
  • Before merge to main

Optional but valuable:

  • When stuck (fresh perspective)
  • Before refactoring (baseline check)
  • After fixing complex bugs

How to Request โ€‹

  1. Get git SHAs:

    bash
    BASE_SHA=$(git rev-parse HEAD~1)
    HEAD_SHA=$(git rev-parse HEAD)
  2. Dispatch reviewer subagent with:

    • What was implemented
    • Plan/requirements reference
    • Base and head SHAs
    • Brief description
  3. Act on feedback:

    • Fix Critical issues immediately
    • Fix Important issues before proceeding
    • Note Minor issues for later
    • Push back if reviewer is wrong (with reasoning)

Part B: Receiving Code Review โ€‹

When to Use โ€‹

When receiving feedback โ€” whether from human reviewers, AI reviewers, or code review subagents.

The Protocol โ€‹

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

Response Framework โ€‹

Feedback TypeResponse
Technically correctFix it. Thank reviewer.
Unclear intentAsk for clarification with specific questions
Technically questionableChallenge with evidence (code, tests, docs)
Stylistic preferenceDiscuss trade-offs, defer to team convention

Red Flags โ€” STOP โ€‹

  • Blindly implementing all suggestions without verification
  • "Performative agreement" โ€” saying yes without understanding
  • Implementing a suggestion that breaks existing tests
  • Making changes you can't justify technically

Anti-Pattern: Performative Agreement โ€‹

โŒ "Good catch! Fixed."  (without verifying it's actually a problem)
โœ… "I verified this: [evidence]. The suggestion is correct because [reason]. Fixed."
โœ… "I investigated this: [evidence]. The current code is correct because [reason]."

Part C: Finishing a Development Branch โ€‹

When to Use โ€‹

When implementation is complete and all tests pass.

The Process โ€‹

  1. Verify current state:

    bash
    npm run test:gate  # All tests must pass
    git status          # Working tree should be clean
  2. Present options to user:

    OptionWhenCommand
    Merge to mainFeature readygit checkout main && git merge feature-branch
    Create PRNeeds team reviewgit push origin feature-branch
    Keep workingMore tasks remainContinue on branch
    Cleanup onlyAbandoned/mergedgit worktree remove path
  3. Execute chosen option

  4. Cleanup:

    • Remove worktree if using cm-git-worktrees
    • Delete feature branch if merged
    • Update task tracking

Rules โ€‹

  • Never merge with failing tests
  • Never force push main/production
  • Always use cm-identity-guard before git push

Integration โ€‹

SkillRelationship
cm-executionReviews after each task in execution
cm-quality-gateTests must pass before finishing branch
cm-identity-guardBefore git push
cm-git-worktreesCleanup worktree after completion

The Bottom Line โ€‹

Review early. Verify feedback. Ship with evidence, not hope.

Open Source AI Agent Skills Framework