google-code-reviewer

Run a Google-style code review workflow for pull requests and changelists. Use when reviewing design, correctness, tests, readability, comments, documentation, review speed, and comment severity (Required/Nit/Optional/FYI). Includes decision branches for large CLs, major design problems, and LGTM with comments.

5 stars

Best use case

google-code-reviewer is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Run a Google-style code review workflow for pull requests and changelists. Use when reviewing design, correctness, tests, readability, comments, documentation, review speed, and comment severity (Required/Nit/Optional/FYI). Includes decision branches for large CLs, major design problems, and LGTM with comments.

Teams using google-code-reviewer should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/google-code-reviewer/SKILL.md --create-dirs "https://raw.githubusercontent.com/agoudbg/Cosmosh/main/.agents/skills/google-code-reviewer/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/google-code-reviewer/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How google-code-reviewer Compares

Feature / Agentgoogle-code-reviewerStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Run a Google-style code review workflow for pull requests and changelists. Use when reviewing design, correctness, tests, readability, comments, documentation, review speed, and comment severity (Required/Nit/Optional/FYI). Includes decision branches for large CLs, major design problems, and LGTM with comments.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

SKILL.md Source

# Google Code Reviewer Skill

Use this skill to perform a structured, high-signal code review inspired by Google Engineering Practices.

## Default Mode
- Strictness profile: `Balanced`.
- Output language: English.
- Comment labeling: `Required`, `Nit`, `Optional`, `FYI`.

## Inputs To Request
- Change context: PR/CL link, summary, related issue/design doc.
- Scope: files to review and any explicit ownership boundaries.
- Risk profile: user-facing impact, security, privacy, data integrity, performance, concurrency.
- Constraints: release deadlines, emergency status, reviewer turnaround expectations.

## Review Standard
- Approve when the change clearly improves overall code health, even if not perfect.
- Do not approve if code health clearly regresses, except true emergencies.
- Prefer evidence, style guide rules, and engineering principles over personal preference.

## Workflow
1. Triage the CL.
- Check if the change should happen at all.
- Read description for clear "what" and "why".
- If direction is wrong, stop early and suggest a better direction.

2. Review high-leverage parts first.
- Find core files and design decisions.
- Send major design concerns immediately without waiting for full pass.

3. Review all assigned code in sequence.
- Cover every human-written line in assigned scope.
- Pull broader file/system context when local diff is insufficient.

4. Evaluate quality dimensions.
- Design: architecture fit, boundaries, extensibility without over-engineering.
- Functionality: correctness, edge cases, failure modes, UX impact.
- Concurrency and safety: race/deadlock risk, state consistency.
- Complexity: readability, maintainability, avoid speculative abstractions.
- Tests: right level, useful assertions, failure sensitivity, maintainable tests.
- Naming/comments/docs: clear names, comments explain why, docs updated when behavior changes.
- Style/consistency: follow style guides first, then local consistency.

5. Write review comments with explicit severity.
- `Required:` must change before approval.
- `Nit:` polish, non-blocking.
- `Optional:` suggestion, author can choose.
- `FYI:` informational only.
- Be respectful, explain reasoning, and compliment strong work when present.

6. Decide review outcome.
- `Approve/LGTM`: no blocking concerns.
- `LGTM with comments`: only minor non-blocking points remain and author is likely to apply them.
- `Request changes`: blocking issues remain.
- For partial reviews, state exactly what was reviewed and what was not.

7. Keep review latency low.
- Target a quick first response and short response loops.
- Maximum normal response time target: within one business day.
- If busy, acknowledge and provide ETA or suggest alternate reviewer.
- For oversized CLs, ask to split into smaller self-contained changes.

## Branching Logic
- Large CL and no fast full review possible: request split first; if unavoidable, provide immediate high-level design feedback.
- Major design flaw found early: send that feedback immediately; avoid deep nits until direction is settled.
- Style-only disagreement not in style guide: treat as preference, usually non-blocking.
- Multiple reviewers: state your reviewed scope explicitly.
- Time-zone delay risk: prefer `LGTM with comments` when only minor issues remain.

## Completion Checklist
- Outcome stated clearly (approve, lgtm-with-comments, or changes requested).
- Blocking vs non-blocking comments are clearly labeled.
- Comments focus on code and rationale, not person.
- Review explicitly covers design, tests, and maintainability.
- Any missing docs/tests are called out.
- If scope was partial, reviewed scope is documented.

## Output Format
- `Verdict:` one-line decision.
- `Blocking Findings:` numbered, each with file/path reference and why it matters.
- `Non-Blocking Suggestions:` numbered and severity-labeled.
- `Positives:` concrete good practices observed.
- `Scope Note:` what you did and did not review.

## Assets
- Use [review output template](./assets/review-output-template.md) when generating structured review feedback.

## Extra References
- Use [eng-practices map](./references/eng-practices-map.md) to select source docs by review situation.

## References
- `./references/eng-practices-full/review/reviewer/index.md`
- `./references/eng-practices-full/review/reviewer/standard.md`
- `./references/eng-practices-full/review/reviewer/looking-for.md`
- `./references/eng-practices-full/review/reviewer/navigate.md`
- `./references/eng-practices-full/review/reviewer/speed.md`
- `./references/eng-practices-full/review/reviewer/comments.md`

Related Skills

google-review-conflict-resolution

5
from agoudbg/Cosmosh

Resolve reviewer-author disagreement using Google-style code review conflict workflow. Use when review comments stall, standards are disputed, or escalation is needed. Produces evidence-based consensus steps, escalation path, and documented final decision.

google-cl-author

5
from agoudbg/Cosmosh

Prepare and iterate changelists/pull requests using Google-style author practices. Use when writing clear CL descriptions, splitting large changes, bundling tests, handling reviewer comments, and keeping dependent CLs build-safe.

web-design-guidelines

5
from agoudbg/Cosmosh

Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

vercel-react-native-skills

5
from agoudbg/Cosmosh

React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.

vercel-react-best-practices

5
from agoudbg/Cosmosh

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

vercel-composition-patterns

5
from agoudbg/Cosmosh

React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.

cosmosh-ui-patterns

5
from agoudbg/Cosmosh

Cosmosh renderer UI design and interaction workflow for implementing, refactoring, or reviewing React/Tailwind/Radix UI work in the Cosmosh Electron app. Use when touching packages/renderer UI, pages, layout, tabs, SSH terminal interactions, SFTP/browser-like workflows, settings surfaces, visual tokens, CSS/Tailwind classes, Radix wrappers, localization, or UI/UX documentation for Cosmosh.

implementing-google-workspace-sso-configuration

16
from plurigrid/asi

Configure SAML 2.0 single sign-on for Google Workspace with a third-party identity provider, enabling centralized authentication and enforcing organization-wide access policies.

implementing-google-workspace-phishing-protection

16
from plurigrid/asi

Configure Google Workspace advanced phishing and malware protection settings including pre-delivery scanning, attachment protection, spoofing detection, and Enhanced Safe Browsing.

implementing-google-workspace-admin-security

16
from plurigrid/asi

Implements comprehensive Google Workspace security hardening including admin console configuration, phishing-resistant MFA enforcement, DLP policies, email authentication (SPF/DKIM/DMARC), OAuth app control, and external sharing restrictions. Activates for requests involving Google Workspace hardening, G Suite security configuration, or cloud office security administration.

google-workspace

16
from plurigrid/asi

Google Workspace MCP integration for Gmail, Drive, Calendar, Docs, Sheets, Slides, Forms, Tasks, and Chat. Use when the user wants to read/send emails, manage files, create/edit documents, schedule events, or interact with any Google Workspace service.

configuring-identity-aware-proxy-with-google-iap

16
from plurigrid/asi

Configuring Google Cloud Identity-Aware Proxy (IAP) to enforce per-request identity verification for Compute Engine, App Engine, Cloud Run, and GKE services using access levels, context-aware policies, and programmatic access with service accounts.