codex-review

Run independent Codex CLI reviews via WSL for plans and code changes.

1|Updated Feb 11, 2026
One-click install
npx skills add https://github.com/dimitri-vs/elevate-agent-skills --skill codex-review-dimitri-vs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/dimitri-vs/elevate-agent-skills/tree/main/codex-review
Command: npx skills add https://github.com/dimitri-vs/elevate-agent-skills --skill codex-review-dimitri-vs

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It reduces the risk of missed bugs, gaps, or weak reasoning by getting an unbiased third-party review of plans, code changes, research, or decisions.

Core Features & Use Cases

  • Independent second-opinion reviews: Reviews can validate implemented features, completed plans, and research/analysis, and can also sanity-check non-code advice and decisions.
  • WSL-backed Codex CLI execution: Runs OpenAI Codex CLI via WSL in the background and returns a concise review.
  • Effort control and safer execution modes: Supports high vs xhigh effort presets and optional sandbox modes (read-only or workspace-write).
  • Action-oriented result handling: Promotes triage of findings into Act On / Note / Noise and emphasizes verifying claims before acting.

Quick Start

Run a background Codex review with your full prompt using the Bash tool with run_in_background: true, for example: cd the codex-review skill directory and execute uv run review.py -w C:\path\to\project "Your full review prompt here".

Frequently Asked Questions about codex-review

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

FAQPage Schema
How do I get an independent second opinion on my code changes?

To get an independent second opinion on your code changes, you can run a background Codex CLI review via WSL. This generates an unbiased third-party review to identify missed bugs, edge cases, or architectural issues in your implemented features.

Can I validate research analysis and non-technical decisions using Codex?

You can validate research analysis and non-technical decisions by running a Codex review. It provides an independent quality check for advice and decisions, reducing the risk of weak reasoning even when conversation history is unavailable.

How do I run a Codex review in the background using WSL?

Run the review in the background by executing the Bash tool with run_in_background set to true in the codex-review directory. Use uv run review.py with your project path and full prompt to capture output and save results.

Does Codex review support different effort levels and sandbox modes?

Codex review supports configurable effort levels using high or xhigh presets. It also offers optional sandbox modes including read-only and workspace-write to ensure safer execution during the review process.

What is the best way to handle findings from an independent code review?

The best way to handle review findings is to triage them into Act On, Note, or Noise categories. You should verify the claims made in the review before taking any action to ensure correctness and completeness.