receiving-code-review Skill: Make Claude Verify Review Feedback

Makes your agent check code review comments against the real codebase before changing anything, and push back when a comment is wrong.

Installs: 228K; on skills.sh
Stars: 310.6K; on GitHub
Updated: Oct 2026; last repo commit

The skill sets a six-step routine for review feedback: read it all, restate it, verify it against the code, decide if it is right for this codebase, reply with agreement or reasoned pushback, then implement one item at a time and test each. Unclear items stop all work until you answer.

It triggers when the agent receives review feedback, before it acts on it. The skill folder holds only SKILL.md. It is one skill in the Superpowers plugin by obra, and it can be installed alone.

Good for

Developers who paste PR review comments into Claude or Codex
Teams tired of replies that open with "You're absolutely right!"
Reviewers whose suggestions may not fit the whole codebase

Install receiving-code-review

Claude Code

Copy this prompt.

Install the receiving-code-review skill in Claude Code. Run this command:

npx skills add https://github.com/obra/superpowers --skill receiving-code-review

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

From its listing on skills.sh. Source.

What's in the skill

When it loads

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

Sections in SKILL.md

Overview
The Response Pattern
Forbidden Responses
Handling Unclear Feedback
Source-Specific Handling
From your human partner
From External Reviewers
YAGNI Check for "Professional" Features
Implementation Order
When To Push Back
Acknowledging Correct Feedback
Gracefully Correcting Your Pushback
Common Mistakes
Real Examples
GitHub Thread Replies

Key facts

Author: obra (Superpowers)
License: MIT, from the crawl of the repo
Installs on skills.sh: 228,020 (Nov 2, 2026)
GitHub stars: 310,635 for the obra/superpowers repo (Nov 2, 2026)
Last repo commit: Oct 29, 2026
What installs: One SKILL.md; no bundled scripts
Needs: One section mentions the gh CLI for replying in GitHub review threads

What it includes

Read, understand, verify, evaluate, respond, then implement one item at a time
Six steps. Each fix is tested on its own.
The response pattern
Bans "You're absolutely right!", "Great point!" and thanks; act or state the fix
The skill treats performative agreement and thanks as mistakes.
Forbidden responses
Stops and asks about unclear items before implementing any of them
Items may be related, so partial understanding gives wrong work.
Handle unclear feedback
Five checks: correct here, breaks anything, why current code, all platforms, full context
Advice from outside reviewers is treated with skepticism and verified first.
Files · External reviewer checks
Greps for real usage before building a "proper" feature a reviewer asked for
If nothing uses it, the agent suggests removing it instead.
Runs code · YAGNI check
Blocking issues first, then simple fixes, then complex ones
Test each fix and check for regressions.
Runs code · Implementation order
Pushes back with technical reasons, and admits it plainly when wrong
No long apologies.
Push back and correct
Replies inside the review comment thread via the gh api, not as a top-level comment
Uses the pull request comment replies endpoint.
Runs code · GitHub thread replies

From the official docs

Permissions

Low risk

The skill can edit code and post pull request replies with your GitHub login. Work on a branch and check replies before they are posted.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It can post replies on GitHub pull requests
When replying to inline comments it uses `gh api` with your login. Ask the agent to show replies before it posts.
It edits code and runs tests
It implements accepted feedback one item at a time. Work on a branch and read the diff.

Technical details

Talks to	GitHub (only if replying to PR threads)

Questions

What does receiving-code-review do?

It changes how your agent handles review comments: it verifies each one against the code, asks about unclear ones, pushes back on wrong ones, and implements one at a time with tests.

How do I install it in Claude Code, the Claude app or Codex?

Use the install box above for this skill alone. The Superpowers README also lists a Claude Code plugin install (/plugin install superpowers@claude-plugins-official) and Codex plugin routes, which give you the whole set.

Does it need an API key?

SKILL.md mentions no key. It mentions the gh CLI for replying in GitHub threads, which uses your own GitHub login.

Is it safe?

It bundles no scripts. It does tell the agent to edit code, run tests and post replies in GitHub pull request threads with gh api, so watch what gets posted and work on a branch.

Will Claude still do what my reviewer asks?

Yes when the comment is right. The skill has it verify first, ask when unclear, and push back only with technical reasons.

Do I need the whole Superpowers plugin?

No. The SKILL.md does not call any other skill, so it can be installed alone. The plugin gives you the rest of the set if you want it.

Information

updated Nov 2, 2026

Repository: obra/superpowers
Folder: skills/receiving-code-review

obra/superpowers skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 2

obra/superpowers receiving-code-review SKILL.md · github.com
obra/superpowers README · github.com