code-review

Guide structured code reviews across correctness, maintainability, security, and performance.

Updated Jan 23, 2026
One-click install
npx skills add https://github.com/datorresb/vibecoding-starter --skill code-review-datorresb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/datorresb/vibecoding-starter/tree/main/.claude/skills/core/code-review
Command: npx skills add https://github.com/datorresb/vibecoding-starter --skill code-review-datorresb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, multi-dimensional guide for code reviews, elevating code quality by focusing on correctness, maintainability, security, and performance.

Core Features & Use Cases

  • Correctness: verify logic, edge cases, and test coverage; identify defects and suggest precise fixes.
  • Maintainability: assess readability, naming, documentation, and modular design; promote consistency.
  • Security & Performance: spot input validation gaps, security risks, and potential performance bottlenecks; suggest mitigations.
  • Feedback & Workflow: deliver constructive, prioritized feedback with actionable steps and severity classifications (CRITICAL, MAJOR, MINOR, NIT) and tie-ins to PR hygiene.

Quick Start

Feed a pull request diff into the code-review skill to generate structured, prioritized feedback across dimensions.

Frequently Asked Questions about code-review

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

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

A structured code review evaluates a pull request diff across correctness, maintainability, security, and performance. Feed the PR diff into the skill to generate prioritized, actionable feedback with severity classifications.

What should a code review checklist cover for software engineering?

A code review checklist should cover logic correctness, edge cases, test coverage, readability, modular design, input validation, and performance bottlenecks. It verifies defects and suggests precise fixes to ensure software quality.

How do I classify code review feedback by severity?

Code review feedback can be classified by severity into CRITICAL, MAJOR, MINOR, and NIT. This prioritization helps deliver constructive feedback and guide PR workflows by highlighting the most urgent issues first.

Can I check security and performance bottlenecks during a code review?

Yes, code reviews can spot input validation gaps, security risks, and potential performance bottlenecks. The skill assesses these dimensions and suggests mitigations to improve overall software correctness and security.

What's the best way to provide actionable code review feedback?

The best way to provide actionable code review feedback is to deliver constructive, prioritized steps tied to PR hygiene. Use severity classifications to structure the review and suggest precise fixes for maintainability issues.