codex

Review code with pass/fail gates and adversarial challenges.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/sobhanashine/nazarato --skill codex-sobhanashine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/sobhanashine/nazarato/tree/main/.claude/skills/codex
Command: npx skills add https://github.com/sobhanashine/nazarato --skill codex-sobhanashine

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, read, write, glob, grep, askuserquestion, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the process of code review, providing an alternative perspective or second opinion, and offers in-depth information when needed.

Core Features & Use Cases

  • Code Review: Automated independent code review via 'codex review' with a pass/fail gate.
  • Challenge: Adversarial mode that challenges your code for robustness.
  • Consultation: Access in-depth information or second opinions for coding questions.
  • Use Case: Use this Skill to review your code, get a second opinion on complex code changes, or consult on any coding-related queries.

Quick Start

To perform a code review, type: codex review.

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 code changes using a CLI?

Automated code review provides an alternative perspective on your changes by running an independent review via the CLI wrapper, operating in plan mode with safe operations to evaluate your code.

What is adversarial mode in code review?

Adversarial challenge mode tests your code for robustness by actively probing weaknesses and edge cases. It operates during plan mode to ensure your logic holds up against alternative perspectives.

How do I perform an automated code review with a pass/fail gate?

You can perform an automated code review with a pass/fail gate by executing 'codex review'. This triggers an independent evaluation that checks your code against safe operation rules and reports a pass or fail outcome.

Can I use bash, grep, and glob operations during a code consultation?

Yes, code consultation supports bash, read, write, glob, grep, and askuserquestion operations. The Skill utilizes these dependencies to navigate, search, and analyze your codebase during the review or consultation process.

Does the code review CLI wrapper support plan mode for safe operations?

Yes, the code review CLI wrapper operates in plan mode with safe operations. It enforces skill invocation rules, askuserquestion formats, and artifact sync to ensure changes are reviewed safely before execution.