skills.rest
24601
24601COMMUNITY
Explore all skills by this author →
Website𝕏@moltar81435GitHub

codex

Review git diffs with adversarial testing and gstack logging.

Updated Mar 28, 2026
Search Tags:#code review#cli#codex#telemetry#git-diff#adversarial-testing#gstack
One-click install
npx skills add https://github.com/24601/gastack-pre-fork --skill codex-24601
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/24601/gastack-pre-fork/tree/main/codex
Command: npx skills add https://github.com/24601/gastack-pre-fork --skill codex-24601

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides an independent, adversarial second opinion on code changes and a consultative channel for coding questions so teams catch issues a single reviewer might miss.

Core Features & Use Cases

  • Three modes: review (diff-based code review with pass/fail gate), challenge (adversarial testing to try to break changes), and consult (interactive Q&A with session continuity).
  • Safety and integration: enforces a filesystem boundary, integrates with the Codex CLI, logs review results to gstack analytics, and supports model reasoning overrides like --xhigh.
  • Use Case: Run a codex review on a feature branch to surface P1/P2 findings, compare results with an earlier Claude review, and log outcomes to the plan file for downstream triage.

Quick Start

Run /codex review to run an independent Codex code review of your current branch diff.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I run an automated code review on a git diff?▼

An AI second-opinion code review provides an independent, adversarial analysis of your code changes to catch issues a single reviewer might miss. It evaluates git diffs and plan files to surface P1/P2 findings for downstream triage.

How does adversarial testing work for code changes?▼

Adversarial testing challenges your code changes to try to break them using AI-driven probes. It operates within an enforced filesystem boundary to safely identify vulnerabilities and edge cases in your git diff.

Can I use an interactive Q&A session for coding questions with session continuity?▼

Yes, consult mode provides an interactive Q&A session with session continuity for coding questions. It acts as a consultative channel to address coding inquiries while maintaining conversational context.

Does the Codex CLI support model reasoning effort overrides?▼

Yes, the Codex CLI integration supports model reasoning effort overrides like the --xhigh flag. This allows you to adjust the analytical depth applied during code review and adversarial testing.

What are the limitations of using an AI second-opinion code review?▼

A key limitation is the enforced filesystem boundary, restricting analysis to local diffs and plan files only. Additionally, review gating and telemetry must be logged to gstack analytics for downstream triage.

Related Skills

Explore More Software Engineering Skills→
YaugourtYaugourt
2

codex

Route code diffs through the Codex CLI for review, challenge, or consult modes.

Community
Advanced
jovermierjovermier
1

quality-severity

Classify issues and code review findings into P1, P2, or P3 severity levels.

Community
Basic
kuroRykuroRy

codex

Run reproducible Codex CLI sessions for code analysis, refactoring, or automated editing.

Community
Advanced
zzxtbetazzxtbeta

codex

Review code diffs against the base branch using the Codex CLI.

Community
Advanced