codex

Evaluate code diffs with Codex CLI in review, challenge, or consult modes.

Updated Mar 24, 2026
One-click install
npx skills add https://github.com/pcaokhai/gradion-response-report-api --skill codex-pcaokhai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/pcaokhai/gradion-response-report-api/tree/main/.claude/skills/gstack/codex
Command: npx skills add https://github.com/pcaokhai/gradion-response-report-api --skill codex-pcaokhai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Codex CLI wrapper provides an independent, blunt second opinion on code changes by orchestrating OpenAI Codex for reviews, adversarial challenges, and consults.

Core Features & Use Cases

  • Supports three modes: review (diff-based gate), challenge (adversarial probing), and consult (session-based questions) to validate code quality and resilience.
  • Integrates with local Git workflows to assess diffs, plan reviews, and acceptance criteria, providing repeatable checks and clear gates.

Quick Start

Install the Codex CLI and run /codex review, /codex challenge, or /codex consult to get a second opinion on your changes.

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 second opinion on my Git diff?

To get an independent second opinion on a Git diff, use the Codex CLI in review mode to evaluate code changes against clear acceptance gates. This orchestrates an adversarial assessment of your local workflow diffs.

What is adversarial code review and how does it validate resilience?

Adversarial code review challenges code resilience by probing diffs for structural weaknesses. Using the challenge mode, the tool provides an independent, blunt critique to validate the robustness of your changes.

How do I use Codex CLI to consult on local Git workflows?

To use Codex CLI for consulting on Git workflows, run the consult mode to ask session-based questions. It provides reasoning emphasis and session continuity for validating plans and follow-up inquiries.

Does the code review tool support session continuity for follow-ups?

Yes, the code review tool supports session continuity for follow-ups. This allows you to maintain context across multiple questions when consulting on code changes and development plans.

Can I use web search during an adversarial code challenge?

Yes, you can use web search during an adversarial code challenge. The tool supports an optional web_search_cached feature to enhance reasoning and context when probing code changes.

What is the best way to validate code quality before merging diffs?

The best way to validate code quality before merging diffs is running an independent review gate. The tool assesses local Git diffs and provides repeatable checks to ensure changes meet acceptance criteria.