Wrap the OpenAI Codex CLI to review git diffs and challenge code.
npx skills add https://github.com/ajsai47/holyclaude --skill codex-ajsai47
Please help me install this Agent Skill. Skill: codex Source: https://github.com/ajsai47/holyclaude/tree/main/skills/team/codex Command: npx skills add https://github.com/ajsai47/holyclaude --skill codex-ajsai47
Claude Code lacks a truly independent second opinion on code quality. This skill wraps the OpenAI Codex CLI to provide an independent, brutally honest review, an adversarial challenge mode to stress-test your code, and a consult mode for follow-up questions with session continuity.
Use the codex skill to review your current branch diff and report any critical findings with a pass or fail verdict.
High-intent search queries and answers about installing and using this skill.
To get an independent AI code review on your git diff, this Skill wraps the OpenAI Codex CLI to analyze your current branch changes against the base branch, providing a pass/fail quality gate verdict on critical findings.
You can use AI to stress-test your code for edge cases by running the adversarial challenge mode, which acts as a highly critical developer to find weaknesses and logic bugs in your implementation before users do.
Yes, this code review tool requires the OpenAI Codex CLI binary to be installed globally via npm on your system before you can execute reviews, challenges, or consult modes within your git-based projects.
Adversarial challenge mode is an independent testing process where an AI attempts to break your code implementation by stress-testing it to discover edge cases, security issues, and architectural weaknesses before deployment.
To maintain session continuity for follow-up architecture questions, the consult mode allows you to ask an independent AI anything about design tradeoffs or debugging problems while preserving the context of previous queries.
A limitation of using a second AI opinion for code quality is that it is scoped exclusively to git diff outputs against a detected base branch, meaning it cannot review uncommitted files or non-git project architectures.
MFS-codeRun OpenAI Codex reviews or adversarial challenges on git diffs.
xvawlValidate Python and TypeScript PRs against linting, typing, tests, and documentation.
stayconnectquickWrap OpenAI Codex CLI to review diffs, run adversarial tests, and consult with session continuity.
joeylife94Review code and run adversarial tests using the OpenAI Codex CLI.
