reviewers

name	reviewers
description	4400+ curated code review prompts from leading open-source repositories
user-invocable	true
version	1.0

Code Reviewers

A collection of 4400+ curated code review guidelines derived from real code review comments in leading open-source repositories.

Attribution

This skill contains content from baz-scm/awesome-reviewers, licensed under Apache 2.0.

Skill Categories

The reviewers are organized into 10 skill domains. Read each file in skills/ for detailed guidance:

Category	File	Focus
AI-Assisted Development	skills/ai-assisted-development.md	Leveraging AI coding assistants while maintaining quality
Code Readability	skills/code-readability.md	Writing clear, maintainable code
Code Refactoring	skills/code-refactoring.md	Improving code structure and design
Data & ML	skills/data-ml.md	Data pipelines and machine learning
DevOps & Cloud	skills/devops-cloud.md	Infrastructure and deployment
Documentation	skills/documentation.md	Code and API documentation
Full-Stack Development	skills/full-stack-development.md	Frontend and backend integration
Secure Coding	skills/secure-coding.md	Security best practices
Team Collaboration	skills/team-collaboration.md	Code review and teamwork
Testing & Debugging	skills/testing-debugging.md	Test strategies and debugging

Reviewer Organization

The 4400+ reviewers in reviewers/ are prefixed by source project. Major projects include:

Prefix	Project	Count
react-*	React	109
kubeflow-*	Kubeflow	62
neon-*	Neon	50
sentry-*	Sentry	48
spring-*	Spring	46
mastodon-*	Mastodon	45
grafana-*	Grafana	43
cypress-*	Cypress	43
bun-*	Bun	37
deno-*	Deno	36
nest-*	NestJS	36
checkov-*	Checkov	38
argo-*	Argo CD	38
posthog-*	PostHog	39
teleport-*	Teleport	39

And 165+ more projects.

Reviewer File Format

Each reviewer follows this format:

---
title: <Review guideline title>
description: <Brief description>
repository: <source OSS repo>
label: <category>
language: <programming language>
comments_count: <number of supporting comments>
repository_stars: <repo popularity>
---

<Detailed guidelines with examples and code blocks>

Usage

Selecting Relevant Reviewers

1. By Language: Filter reviewers by the language field in their frontmatter

   grep -l "language: TypeScript" reviewers/*.md
   

2. By Project/Framework: Use project prefixes to find relevant patterns

   • React apps: react-*, next-*
   • Python: airflow-*, kubeflow-*, boto3-*
   • Go: kubernetes-*, teleport-*, argo-*
   • Rust: deno-*, bun-*, workerd-*

3. By Category: Use the label field

   • Code Style
   • Performance
   • Security
   • Testing
   • Documentation

Applying Reviewers to Code Review

When reviewing code:

1. Identify the tech stack from the files being reviewed
2. Select 5-10 relevant reviewers based on:
   • Programming language
   • Framework/library in use
   • Type of change (feature, refactor, bugfix)
3. Read each selected reviewer for specific guidelines
4. Apply the patterns when commenting on the code

Integration with Differential Review

When using the differential-review skill:

1. First identify changed files and their languages
2. Load relevant reviewers for those languages/frameworks
3. Use reviewer guidelines to inform your review comments

Quick Reference by Technology

TypeScript/JavaScript:

• react-*, next-*, angular-*, nest-*, bun-*, deno-*

Python:

• airflow-*, kubeflow-*, boto3-*, checkov-*, prowler-*

Go:

• kubernetes-*, teleport-*, argo-*, grafana-*

Rust:

• deno-* (core), bun-* (native), workerd-*

Infrastructure/DevOps:

• kubernetes-*, argo-*, grafana-*, checkov-*, prowler-*

Security:

• checkov-*, prowler-*, teleport-*, sentry-*