code-review

Audit source code for security and quality issues, outputting a JSON report.

482|100|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/agulli/atlas-agents --skill code-review-agulli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/agulli/atlas-agents/tree/main/ch09_agent_skills/skills/code-review
Command: npx skills add https://github.com/agulli/atlas-agents --skill code-review-agulli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured approach to auditing source code for security vulnerabilities and quality issues before merging.

Core Features & Use Cases

  • Security pass: OWASP Top 10 vulnerabilities are checked in a prioritized order.
  • Logic pass: Detect unhandled exceptions, race conditions, off-by-one errors, and mutable defaults.
  • Style pass: Flag readability and maintainability concerns such as long functions and deep nesting.
  • Report generation: Produce a structured JSON report with severity, file, line, and description.

Quick Start

Load the target codebase and run the code-review skill to produce a structured JSON findings report.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate a security and quality audit of source code before merging?

To automate a source code security and quality audit, load your target codebase and run the code-review skill to execute structured checks across OWASP Top 10 vulnerabilities, logic errors, and style concerns, generating a structured JSON findings report.

Can I check for OWASP Top 10 vulnerabilities and race conditions during a pull request audit?

Yes, pull request audits can check for OWASP Top 10 vulnerabilities and race conditions. The skill performs a prioritized security pass for vulnerabilities and a logic pass to detect race conditions, unhandled exceptions, and off-by-one errors.

What is the best way to generate a structured vulnerability assessment report for my codebase?

The best way to generate a structured vulnerability assessment report is to run an automated code audit. This skill outputs a JSON report detailing the severity, file, line number, and description of each security vulnerability and quality issue found.

Does this static analysis tool flag readability and maintainability concerns like long functions?

Yes, this static analysis tool flags readability and maintainability concerns. During its style pass, it identifies style issues such as long functions and deep nesting to help enforce quality gates before merging.

What specific logic errors are detected during a code review?

A code review detects specific logic errors including unhandled exceptions, race conditions, off-by-one errors, and mutable defaults. These issues are identified during the logic pass and documented in the final JSON report.

What format does the code review audit output for vulnerability findings?

The code review audit outputs vulnerability findings in a structured JSON format. This report includes the severity level, affected file, line number, and a description of each detected security vulnerability or quality issue.