review-bugs

Detect logic errors, race conditions, and edge cases in code changes.

4|1|Updated Jan 11, 2026
One-click install
npx skills add https://github.com/doodledood/codex-workflow --skill review-bugs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-bugs
Source: https://github.com/doodledood/codex-workflow/tree/main/skills/review-bugs
Command: npx skills add https://github.com/doodledood/codex-workflow --skill review-bugs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audits code changes for bugs, logic errors, race conditions, and edge cases without modifying the source, delivering a concise, actionable report.

Core Features & Use Cases

  • Read-only code review that never alters code
  • Scoped analysis based on changed files or explicit paths
  • Generates structured bug reports with clear severity and categories

Quick Start

Provide the files or diff scope to review, and ask for a bug review report.

Frequently Asked Questions about review-bugs

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

FAQPage Schema
How do I find logic errors and race conditions in a pull request?

To find logic errors and race conditions in a pull request, perform a read-only bug review on the diff against origin/main to generate a structured categorized report without modifying code.

What is the best way to do a read-only code review for edge cases?

A read-only code review for edge cases involves auditing changed files or explicit paths to identify bugs and deliver a concise actionable report without altering the source code.

Can I use this bug detection approach on explicit file paths instead of a diff?

Yes, you can use bug detection on explicit file paths instead of a diff, applying scoped analysis to identify logic errors and generate a categorized bug report for post-implementation review.

Does this code review method modify my source code during bug hunting?

No, this code review method is strictly read-only and never modifies your source code during bug hunting, ensuring the repository remains untouched while producing a structured reliability report.

How do I generate a structured bug report after implementing new code?

To generate a structured bug report after implementing new code, run a scoped bug review on the changed files to identify logic errors, race conditions, and edge cases categorized by severity.

When should I use a read-only bug review instead of an automated code fix?

Use a read-only bug review instead of an automated code fix when you need to safely identify logic errors and edge cases for PR checks without risking unintended modifications to the source code.