Skip to content

stratuslabs/skill-code-review

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

4 Commits
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ”¬ skill-code-review

Systematic code review patterns for AI agents.

A skill for StratusOS β€” the AI operating system for Mac.


What it does

Teaches any agent how to review code like a senior engineer β€” security vulnerabilities, performance issues, maintainability concerns, and style consistency. Includes structured output formats for GitHub PR comments, severity ratings, and actionable fix suggestions.

Installation

# From StratusOS Marketplace (recommended)
# Open StratusOS β†’ Marketplace β†’ Skills β†’ Code Review β†’ Install

# Or manual:
git clone https://github.com/stratuslabs/skill-code-review.git ~/.stratusos/skills/code-review

What's included

SKILL.md                β€” Core review methodology
templates/
  pr-review.md          β€” GitHub PR review format (severity + findings + suggestions)
  security-audit.md     β€” Security-focused review checklist
  performance-review.md β€” Performance bottleneck analysis
  refactor-plan.md      β€” Refactoring recommendations with effort estimates

Review categories

  • Security β€” injection, auth bypass, data exposure, secrets in code
  • Performance β€” N+1 queries, memory leaks, unnecessary allocations
  • Correctness β€” race conditions, edge cases, error handling
  • Maintainability β€” naming, complexity, dead code, missing tests
  • Style β€” consistency, formatting, idiomatic patterns

How skills work

Skills are pure knowledge β€” no binaries, no APIs. When loaded, the skill's SKILL.md and templates are injected into the agent's context. The agent learns the review methodology and applies it to any codebase.

Requirements

  • StratusOS v0.2.0+


Built for StratusOS β€” your AI operating system.
GitHub Β· Discord

Made by Stratus Labs

About

πŸ”¬ Systematic code review patterns β€” security, performance, correctness

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors