codex

Run Codex CLI and Claude's code-reviewer subagent for parallel code review.

1|Updated Jun 13, 2017
One-click install
npx skills add https://github.com/RoyMcCrain/dotsfile --skill codex-roymccrain
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/RoyMcCrain/dotsfile/tree/main/claude/skills/codex
Command: npx skills add https://github.com/RoyMcCrain/dotsfile --skill codex-roymccrain

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Developers need fast, reliable code reviews, and relying on a single tool can miss insights; this skill provides automated review by combining Codex CLI and Claude.

Core Features & Use Cases

  • Parallel Review: Runs Codex MCP in a sandbox and Claude's code-reviewer subagent simultaneously.
  • Result Merging: Consolidates matching, supplemental, and conflicting findings into a single report.
  • Typical Use: Quickly assess a new pull request or script for style, bugs, and best‑practice compliance.

Quick Start

Run /codex Review app.py for potential bugs and style issues.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I run automated code reviews for potential bugs and style issues?

Automated code reviews are performed by running Codex CLI alongside Claude's code-reviewer subagent in parallel. This combination analyzes source files for style, correctness, and best practices, consolidating findings into a single report.

How does parallel AI code analysis work with Codex and Claude?

Parallel AI code analysis works by simultaneously executing Codex MCP in a sandbox read-only mode and Claude's code-reviewer subagent. The system then merges matching, supplemental, and conflicting findings into a unified review report.

Do I need Codex MCP sandbox access to review source files?

Yes, you need Codex MCP access configured in sandbox read-only mode. This environment setup is required to execute the Codex CLI portion of the parallel code review process safely.

What is the best way to assess a new pull request for best-practice compliance?

Assessing a pull request for best-practice compliance is best handled by combining multiple reviewers. Running Codex and Claude in parallel captures more insights than a single tool, preventing missed bugs or style violations.

How do I consolidate conflicting findings from multiple code review tools?

Conflicting findings from multiple code review tools are consolidated automatically by this Skill. It merges results from Codex and Claude into a single report, categorizing matching, supplemental, and conflicting issues.