skills.rest
tgmarinho
tgmarinhoCOMMUNITY
Explore all skills by this author →
Website𝕏@tgmarinhoGitHub

codex

Runs OpenAI Codex CLI for independent code review, adversarial challenge, and consult sessions.

Updated May 15, 2026
Search Tags:#code review#codex#adversarial testing#second opinion#diff review#consult#openai cli
One-click install
npx skills add https://github.com/tgmarinho/canetaco --skill codex-tgmarinho
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/tgmarinho/canetaco/tree/main/.claude/skills/codex
Command: npx skills add https://github.com/tgmarinho/canetaco --skill codex-tgmarinho

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Getting a second opinion on code changes usually means waiting for a human reviewer or trusting the same model that wrote the code. This Skill wraps the OpenAI Codex CLI so you can get an independent, cross-model review of your diff, an adversarial attempt to break your code, or a consultative Q&A session with continuity across follow-ups. ## Core Features & Use Cases - Code Review Mode: Runs codex review on your diff as an independent pass/fail gate before shipping. - Challenge Mode: Adversarially tries to break your code, surfacing edge cases and failure modes you missed. - Consult Mode: Ask Codex anything with session continuity, so follow-up questions keep prior context. - Use Case: Before merging a branch, invoke a codex review to get a cross-model verdict on the diff, then run challenge mode on the riskiest function to hunt for bugs the first pass missed. ## Quick Start Ask the assistant to run a codex review on the current branch diff and report any blocking findings.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I get a second opinion on a code diff before merging?▼

Invoke the codex review mode, which runs the OpenAI Codex CLI against your current diff and returns an independent pass/fail verdict. This gives you a cross-model check that is separate from the model that wrote the code.

What is adversarial code challenge mode?▼

Challenge mode instructs Codex to actively try to break your code by hunting for edge cases, failure modes, and unhandled inputs. It is useful for stress-testing risky functions before they ship.

Does codex consult support follow-up questions?▼

Yes, consult mode maintains session continuity, so follow-up questions retain the context of earlier exchanges. You can ask an initial question and then drill into details without restating the problem.

What are the requirements to run codex review?▼

You need the OpenAI Codex CLI installed and available on your PATH, plus Bash execution permission. The skill also uses Read, Write, Glob, Grep, and AskUserQuestion for its workflow gates.

When should I use codex review instead of a normal code review?▼

Use it when you want an independent voice that did not write the code, such as before merging a branch or when a change touches critical logic. It complements rather than replaces human review.

Related Skills

Explore More Software Engineering Skills→
MINT-IAMINT-IA

gsd-pr-branch

Filter planning commits from a Git branch to produce a clean PR branch.

Official
Intermediate
ShaheerKhawajaShaheerKhawaja
8

ship-safe

Chain self-evaluation, review, and ship steps to validate code before deployment.

Community
Advanced
erfanzarerfanzar
27

requesting-code-review

Automate pre-commit code review with static checks, linting, and auto-fixes.

Community
Advanced
eklyukineklyukin

codex

Obtains independent second opinions from OpenAI Codex CLI for planning, code review, and implementation review.

Community
Intermediate