codex-review

Automate iterative code review with Codex CLI and Claude Code.

Updated Dec 30, 2025
One-click install
npx skills add https://github.com/lihs-ie/lambars --skill codex-review-lihs-ie
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/lihs-ie/lambars/tree/main/.claude/skills/codex-review
Command: npx skills add https://github.com/lihs-ie/lambars --skill codex-review-lihs-ie

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the iterative code review process, ensuring code quality and adherence to specifications before commits or merges, by leveraging AI for both review and suggested fixes.

Core Features & Use Cases

  • Iterative Review Loop: Automates the cycle of review, AI-driven code correction, and re-review until a satisfactory state (ok: true) is reached.
  • Scalable Review Strategy: Adapts review depth (diff, arch, cross-check) based on the scale of code changes (small, medium, large).
  • Use Case: Integrate this Skill into your Git workflow to automatically perform a thorough, AI-assisted code review before creating a pull request, catching potential issues early and reducing manual review burden.

Quick Start

Use the codex-review skill to review the changes in the current git diff.

Frequently Asked Questions about codex-review

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

FAQPage Schema
How do I automate code reviews before a git commit?

To automate code reviews before a git commit, you can use an iterative AI assistant loop that evaluates git diff changes, suggests corrections, and re-reviews until issues are resolved.

How does an AI assistant handle iterative code correction?

Iterative code correction works by cycling through AI-driven reviews, applying suggested fixes, and re-evaluating the code until a satisfactory state is reached or a maximum iteration count is hit.

Can I scale AI code review depth based on the size of git changes?

Yes, AI code review depth scales with the size of changes by adapting the review strategy to focus on diff, architecture, or cross-checking for small, medium, and large code changes respectively.

What is the JSON output format for AI code review feedback?

AI code review feedback is structured using a specific JSON schema that includes fields for severity, category, and actionable recommendations to clearly define required corrections.

How are execution errors and timeouts handled during automated git workflow reviews?

Automated git workflow reviews handle execution errors and timeouts by terminating the iterative cycle safely, ensuring the process stops gracefully when limits are reached.