google-code-reviewer

Review pull requests with Google-style structured findings and verdicts.

7|1|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/agoudbg/Cosmosh --skill google-code-reviewer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: google-code-reviewer
Source: https://github.com/agoudbg/Cosmosh/tree/main/.agents/skills/google-code-reviewer
Command: npx skills add https://github.com/agoudbg/Cosmosh --skill google-code-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

This skill removes uncertainty and inconsistency from code reviews by providing a structured, high-signal review workflow that highlights blocking issues, prioritizes high-leverage feedback, and labels findings by severity so teams can move forward confidently.

Core Features & Use Cases

  • Structured triage and high-leverage review: quickly determine whether a change should proceed, request a split for oversized CLs, or surface major design concerns early.
  • Multi-dimensional evaluation: assess design, correctness, tests, concurrency, performance, naming/comments/docs, and style while producing labeled findings (Required, Nit, Optional, FYI).
  • Practical output and templates: produces a clear Verdict, numbered Blocking Findings, Non-Blocking Suggestions, Positives, and a Scope Note using a reusable review-output-template; ideal for reviewing GitHub/GitLab/Phabricator PRs or internal changelists.

Quick Start

Run a Google-style code review on the provided PR link, focusing on design, correctness, tests, readability, and label each finding as Required, Nit, Optional, or FYI.

Frequently Asked Questions about google-code-reviewer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform a structured code review on a pull request?

Perform a structured code review by evaluating design, correctness, tests, readability, comments, and documentation to produce a verdict, numbered blocking findings, non-blocking suggestions, and a scope note. This approach assesses PR context, changed files, and risk profile to deliver high-leverage feedback.

What is Google-style code review triage?

Google-style code review triage is the process of quickly determining whether a change should proceed, requesting a split for oversized changelists, or surfacing major design concerns early. It prioritizes high-signal feedback and highlights blocking issues to ensure teams move forward confidently.

Can I use this Google-style review workflow for GitLab and Phabricator changelists?

Yes, this Google-style review workflow is applicable for reviewers evaluating GitHub, GitLab, or Phabricator pull requests and changelists. It handles small to large changelists across these platforms, requiring triage, labeled findings, and high-leverage feedback based on the PR context.

How are code review findings labeled and categorized by severity?

Code review findings are labeled by severity as Required, Nit, Optional, or FYI. This multi-dimensional evaluation categorizes blocking findings and non-blocking suggestions separately, allowing reviewers to assess design, correctness, concurrency, performance, naming, comments, docs, and style effectively.

What is the best way to review oversized pull requests?

The best way to review oversized pull requests is through structured triage that quickly surfaces major design concerns and requests a split for oversized changelists. This high-leverage review workflow assesses scope and risk profile early, ensuring consistent evaluation of large changes.

What output format does a structured code review produce?

A structured code review produces a clear Verdict, numbered Blocking Findings, Non-Blocking Suggestions, Positives, and a Scope Note. This reusable review-output-template format ensures high-signal feedback that highlights blocking issues and prioritizes high-leverage improvements for the team.