Codex Review

Name: Codex Review
Author: alinaqi

🌐Community

by alinaqi · vlatest · Repository

Analyzes backend code for potential bugs, security vulnerabilities, and adherence to coding standards.

Install on your platform

We auto-selected Claude Code based on this skill’s supported platforms.

Run in terminal (recommended)

terminal

claude mcp add alinaqi-codex-review npx -- -y @trustedskills/alinaqi-codex-review

Or manually add to ~/.claude/settings.json

~/.claude/settings.json

{
  "mcpServers": {
    "alinaqi-codex-review": {
      "command": "npx",
      "args": [
        "-y",
        "@trustedskills/alinaqi-codex-review"
      ]
    }
  }
}

Requires Claude Code (claude CLI). Run claude --version to verify your install.

About This Skill

What it does

The Codex Review skill leverages OpenAI's GPT-5.2-Codex model to perform specialized code reviews for backend projects. It analyzes code for potential bugs, security vulnerabilities (like SQL injection), and adherence to coding standards. The tool provides structured output in JSON format and integrates with GitHub by adding review comments directly within pull requests. It can also be used in headless mode for CI/CD automation.

When to use it

Before opening a Pull Request: Review changes against the base branch (e.g., main) to identify issues before merging.
Before committing code: Analyze staged, unstaged, and untracked changes to catch problems early.
Analyzing specific commits: Examine particular commit SHAs from your project's history.
Automated CI/CD pipelines: Integrate Codex Review into automated workflows for continuous code quality checks.

Key capabilities

Uses OpenAI’s GPT-5.2-Codex, specifically trained for code review.
Detects bugs, security flaws, and style issues (reported 88% detection rate on LiveCodeBench).
Provides structured output in JSON schema format.
Integrates natively with GitHub to add comments directly to pull requests.
Supports headless mode for CI/CD automation.
Offers review presets for common use cases (base branch, uncommitted changes, specific commits).
Allows custom instructions to focus the review on particular areas (e.g., security vulnerabilities).

Example prompts

/review - Initiates an interactive code review session.
"Review against base branch" - Compares current changes with the main branch.
"Focus on security vulnerabilities" - Instructs Codex to prioritize identifying potential security issues.

Tips & gotchas

Prerequisites: Requires Node.js version 22 or higher and either a ChatGPT Plus/Pro/Team/Edu/Enterprise subscription or an OpenAI API key.
Authentication: Ensure you are properly authenticated with either your ChatGPT account or by setting the OPENAI_API_KEY environment variable.
Interactive Mode: The initial interaction is through a text-based user interface (TUI).

View Repository →

TrustedSkills Verification

Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.

Security Audits

Gen Agent Trust Hub	Pass
Socket	Pass
Snyk	Pass

Details

Version: vlatest
License
Author: alinaqi
Installs: 65

Repository (canonical source) →

🌐 Community

Passed automated security scans.