codex

Wrap the Codex CLI to review code diffs with pass/fail gates.

Updated Feb 5, 2026
One-click install
npx skills add https://github.com/CaltexBevo/ihe-pulse --skill codex-caltexbevo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/CaltexBevo/ihe-pulse/tree/main/.claude/skills/gstack.bak/codex
Command: npx skills add https://github.com/CaltexBevo/ihe-pulse --skill codex-caltexbevo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Wraps the OpenAI Codex CLI to provide an independent, blunt second opinion on your code changes, helping you validate ideas and surface issues quickly.

Core Features & Use Cases

  • Review mode: code diffs evaluated with a pass/fail gate to surface quality issues.
  • Challenge mode: adversarial testing that attempts to break your changes and reveal weaknesses.
  • Consult mode: ask Codex questions with session continuity for follow-ups.

Quick Start

Run codex review on the current diff to get an independent second opinion.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I get an independent AI code review on my current diff?

To get an independent AI code review on your diff, you can use the review mode to evaluate changes against a pass/fail quality gate, quickly surfacing potential issues.

What is adversarial code testing and how does it reveal weaknesses?

Adversarial code testing attempts to actively break your changes to reveal hidden weaknesses, utilizing a challenge mode to rigorously probe your code for structural flaws.

Do I need the Codex CLI installed to run code review and consult commands?

Yes, you need the Codex CLI installed to run these code review and consult commands, because the system operates as a wrapper that interfaces directly with that CLI.

Can I ask follow-up questions during an AI coding consultation session?

You can ask follow-up questions during an AI coding consultation by using the consult mode, which maintains session continuity to handle general inquiries and subsequent prompts.

When should I use an adversarial prompt instead of a standard code review?

You should use an adversarial prompt instead of a standard code review when you need to aggressively challenge your logic and surface vulnerabilities, rather than just checking basic quality gates.