skills.rest
curielmc
curielmcCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

codex

Run Codex CLI reviews, challenges, or consultations on code changes.

Updated Feb 27, 2026
Search Tags:#workflow#security#codex#code-quality#ai-assistant#code-review#second-opinion
One-click install
npx skills add https://github.com/curielmc/beat-SPY --skill codex-curielmc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/curielmc/beat-SPY/tree/main/.claude/skills/gstack/codex
Command: npx skills add https://github.com/curielmc/beat-SPY --skill codex-curielmc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Codex provides an independent, brutally honest second opinion on your code changes by wrapping the Codex CLI to review, challenge, or consult on your work. It helps catch edge cases, design flaws, and potential vulnerabilities that you might miss in standard reviews.

Core Features & Use Cases

  • Independent analysis: Run Codex in review, adversarial challenge, or consult mode to surface gaps and improvements.
  • Gate-style evaluation: Outputs gate verdicts (PASS/FAIL) with token cost and traceability to the underlying diffs.
  • Session continuity & telemetry: Maintains a conversation thread across prompts, with optional telemetry data for improvement.
  • Plan-aware prompting: Integrates with project plans to align feedback with planning workflows.

Quick Start

Ask Codex to review or challenge your current branch against the base, or consult on a plan, and review the full output.

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 code review on my branch changes?▼

You can get an independent code review by asking Codex to review your branch changes against the base. It executes the Codex CLI to detect edge cases, design flaws, and vulnerabilities, returning a gate verdict with token cost and verification status.

What is an adversarial code review and how does it challenge code changes?▼

Adversarial code review challenges your changes by running an adversarial prompt against your branch diff. This independent analysis surfaces design gaps, edge cases, and improvements that standard reviews might miss.

Can I use AI to consult on a project plan and align feedback with my workflow?▼

Yes, you can use Codex in consult mode to consult on a plan. It supports plan-aware prompting to align its independent feedback directly with your existing project planning workflows.

Does the code review output a pass or fail gate verdict?▼

Yes, the code review outputs a gate-style evaluation with PASS or FAIL verdicts. It includes token cost and traceability to the underlying diffs while preserving the original Codex reasoning traces.

How do I maintain session continuity across multiple code review prompts?▼

You maintain session continuity across multiple code review prompts using the built-in conversation thread feature. It preserves context across prompts while supporting streaming output and optional telemetry data.

Related Skills

Explore More Software Engineering Skills→
mei28mei28
3

codex-review

Runs a read-only Codex review on your working-tree diff and outputs a verdict with fixes.

Community
Intermediate
FrekiManagarmFrekiManagarm

Verification & Quality Assurance

Compute truth scores and run correctness, security, performance, and documentation checks with JSON reports and Git rollback.

Community
Advanced
haidonglethqbhaidonglethqb

project-management

Coordinate multi-plan project management by synchronizing Claude Tasks with plan files.

Community
Advanced
icartshicartsh
4

code-analyze

Analyze .NET solutions for code quality, security, and dependency issues.

Community
Advanced