receiving-code-review

name	receiving-code-review
description	Use when you have just received code review feedback (human or agent) and are about to act on it — verify each item against the codebase before implementing, and push back with technical reasoning instead of performative agreement
metadata	{"category":"development","agent_type":"general-purpose","origin":"adapted from obra/superpowers receiving-code-review"}

Receiving Code Review

Receiving code review requires technical evaluation, not emotional performance. This skill governs what happens between getting feedback and typing the first edit — not how to produce a review.

Core principle: Verify before implementing. Ask before assuming. Technical correctness over social comfort.

When to Use

• You just received review comments (from a human, code-review, review, or pr-multi-perspective-review) and are about to act on them
• Feedback includes multiple items and some are unclear or only partially understood
• A suggestion seems plausible but you have not yet checked it against this codebase
• You pushed back earlier and now have evidence the reviewer was actually right

When NOT to Use

Instead of receiving-code-review	Use
You are producing the review, not receiving it	code-review, review, or pr-multi-perspective-review
You are checking a delegated implementation against a task spec after it landed	implementation-review
Feedback is already fully verified and unambiguous	Just implement it — this skill governs the reception gap, not every edit

The Response Pattern

1. READ: read the complete feedback without reacting or starting to type a fix
2. UNDERSTAND: restate each requirement in your own words, or ask
3. VERIFY: check the claim against the actual codebase (grep, view, run tests)
4. EVALUATE: is this technically sound for THIS codebase, not codebases in general?
5. RESPOND: technical acknowledgment or reasoned pushback — never gratitude theater
6. IMPLEMENT: one item at a time, testing each before moving to the next

Forbidden Responses

Never open with:

• "You're absolutely right!"
• "Great point!" / "Excellent feedback!"
• Any gratitude expression ("Thanks for catching that")
• "Let me implement that now" — before verification has actually happened

Instead:

• Restate the technical requirement
• Ask a specific clarifying question
• Push back with technical reasoning if the suggestion is wrong for this codebase
• Just do the work — the diff is the acknowledgment

If you catch yourself about to write "Thanks" or "You're right" before checking anything, delete it and state the verified fix instead.

Handling Unclear or Partial Feedback

IF any item in a multi-item review is unclear:
    STOP — do not implement any of the items yet
    ASK for clarification on the unclear items specifically

Items in a review batch are often related. Implementing the clear items while silently guessing at the unclear ones produces a partially-wrong result that looks complete.

Example:

Reviewer: "Fix items 1–6." You understand 1, 2, 3, 6. Items 4 and 5 are ambiguous.

Wrong	Right
Implement 1, 2, 3, 6 now; ask about 4, 5 later	"I understand items 1, 2, 3, and 6. I need clarification on 4 and 5 before proceeding — [specific question]."

Verification Before Implementation

Before implementing any single suggestion, check:

1. Is this technically correct for this codebase (not codebases in general)?
2. Does it break existing, tested functionality?
3. Is there a reason the current implementation looks the way it does (check git blame / commit message / adjacent comments before assuming it's a mistake)?
4. Does it hold across the platforms, runtimes, or versions this repo actually targets?
5. Does the reviewer appear to have the full context (or is a detail likely missing on their side)?

IF the suggestion seems wrong:
    push back with technical reasoning, cite the codebase evidence

IF you cannot verify it without more information:
    say so explicitly — "I can't verify this without X. Should I investigate, ask, or proceed?"

IF it conflicts with a prior architectural decision by the project owner:
    stop and raise that conflict before implementing either side

YAGNI Check for "Do It Properly" Suggestions

When a reviewer suggests generalizing or "implementing properly":

grep the codebase for actual current usage of the affected code path

IF unused: propose removing it (YAGNI) instead of hardening it
IF used: implement the suggested generalization properly

Implementation Order

1. Resolve all unclear items first (ask before touching code)
2. Then implement in this order:
   a. Blocking issues (breaks, security)
   b. Simple fixes (typos, imports, naming)
   c. Complex fixes (refactoring, logic changes)
3. Test each fix individually before moving to the next
4. Re-run the full suite once at the end to catch cross-item regressions

When to Push Back

Push back — with technical reasoning, not defensiveness — when the suggestion:

• Would break existing, tested functionality
• Reflects a reviewer who is missing repository-specific context
• Violates YAGNI (adds unused generality)
• Is technically incorrect for this stack, runtime, or platform
• Exists for a documented legacy/compatibility reason
• Conflicts with a prior architectural decision that needs a human, not a silent override

Reference working tests or code when pushing back. If you are uncomfortable disagreeing out loud, name that tension explicitly rather than silently complying.

Correcting Your Own Pushback

If you pushed back and it turns out you were wrong:

✅ "Verified — you're correct. I initially missed [X]. Fixing now."
❌ A long apology
❌ Defending why you pushed back in the first place
❌ Over-explaining

State the correction factually and move on to the fix.

Common Rationalizations

Rationalization	Reality
"The reviewer is senior/the tool is confident, so it's probably right."	Confidence is not verification. Check the codebase before agreeing or disagreeing.
"Agreeing quickly shows I'm responsive."	Performative agreement without verification produces wrong fixes dressed as fast ones.
"I'll implement the clear items and figure out the rest as I go."	Ambiguous items are often coupled to the clear ones; guessing compounds the error.
"Pushing back looks defensive."	Reasoned, evidence-based pushback is a technical contribution, not an excuse.

Red Flags

• The first line of your response is gratitude or agreement before you have opened a single file
• You are implementing item 4 while still unsure what item 3 actually meant
• All feedback items are being applied in review order rather than blocking-issues-first
• You cannot point to a specific grep, test run, or file read that verified a suggestion
• A pushback response has no cited codebase evidence — it is just a preference

Verification

• Every unclear item was clarified before any implementation began
• Each suggestion was checked against this codebase, not just accepted on reviewer authority
• No response opens with unearned gratitude or agreement
• Pushback (if any) cites specific code, tests, or history
• Blocking issues were implemented before cosmetic ones
• Each fix was tested individually, then the full suite re-run at the end

See Also

• code-review — producing a single-pass technical review
• review — producing a two-axis (Standards/Spec) review
• pr-multi-perspective-review — producing a six-lens PR review
• implementation-review — verifying a delegated implementation against its task spec after it lands