codex-review

Automate quality-gate review for Claude Code changes with read-only sandbox execution.

1|Updated Aug 10, 2024
One-click install
npx skills add https://github.com/takumi12311123/dotfiles --skill codex-review-takumi12311123
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/takumi12311123/dotfiles/tree/main/.claude/skills/codex-review
Command: npx skills add https://github.com/takumi12311123/dotfiles --skill codex-review-takumi12311123

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the mandatory quality-gate review for Claude Code changes, ensuring blockers are resolved before any user-facing steps.

Core Features & Use Cases

  • Read-only Codex-based review gate that runs before user-facing steps.
  • Iterates until ok:true, offering automatic fixes for blocking issues when enabled.
  • Produces structured JSON outputs for downstream tooling and audits.

Quick Start

Trigger the codex-review gate on a change set to automatically validate blockers and halt progression until all blocking issues are resolved.

Frequently Asked Questions about codex-review

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

FAQPage Schema
How do I automate a code review gate before presenting changes?

A code review gate automates pre-commit and pre-PR validation by running a read-only sandbox review that iterates until all blocking issues are resolved. It outputs a structured JSON report to ensure deterministic evaluation before user-facing steps.

What is a read-only sandbox code review and how does it work?

A read-only sandbox code review executes automated quality-gate checks without modifying the source environment. It evaluates development workflows and produces structured JSON outputs for downstream tooling, ensuring blockers are caught safely before progression.

How do I integrate pre-commit validation into my development workflow?

You integrate pre-commit validation by applying an automated quality-gate to your change set. The review runs iteratively, automatically fixing blocking issues if enabled, and halts progression until the evaluation returns an ok:true status.

Can I auto-fix blocking issues during an automated code review?

Yes, you can auto-fix blocking issues during an automated code review when the auto-fix setting is enabled. The quality-gate iterates through validation checks, automatically resolving blocking items until the review returns ok:true.

Does code review automation support structured JSON output for audits?

Yes, code review automation supports structured JSON output for audits. The quality-gate generates deterministic reports detailing blocking issues, which can be directly consumed by downstream tooling for compliance and tracking.

What are the limitations of automated quality-gate reviews?

The main limitation of automated quality-gate reviews is that progression halts entirely until all blocking issues are resolved and ok:true is returned. If auto-fix is not enabled, manual intervention is required to clear blockers before proceeding.