skills.rest
DmitriyYukhanov
DmitriyYukhanovCOMMUNITY
Explore all skills by this author →
Website𝕏@dmitriy_focusGitHub

cross-review

Orchestrate concurrent code reviews between Claude agents and Codex CLI.

6|Updated Jan 27, 2026
Search Tags:#quality assurance#code review#codex#automated testing#architecture review#claude agents
One-click install
npx skills add https://github.com/DmitriyYukhanov/claude-plugins --skill cross-review
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-review
Source: https://github.com/DmitriyYukhanov/claude-plugins/tree/main/plugins/cross-review/skills/cross-review
Command: npx skills add https://github.com/DmitriyYukhanov/claude-plugins --skill cross-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates a rigorous, multi-model review process for code, plans, and architecture, ensuring higher quality by leveraging both Claude and Codex CLI.

Core Features & Use Cases

  • Simultaneous Review: Claude agents and Codex CLI review artifacts concurrently.
  • Automated Triage & Fix: Identifies agreements, disagreements, and auto-fixable issues, then applies fixes using the best available skill.
  • Use Case: Before merging a critical feature, use this Skill to have Claude and Codex CLI independently review the code, identify potential bugs or architectural flaws, and automatically fix any clear-cut issues, presenting only the critical disagreements for your decision.

Quick Start

Use the cross-review skill to perform a dual review on the current code changes.

Frequently Asked Questions about cross-review

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

FAQPage Schema
How do I run a multi-model code review using Claude agents and Codex CLI?▼

To run a multi-model code review, this Skill orchestrates an autonomous review-fix loop where Claude agent teams and the Codex CLI concurrently evaluate your code, plans, and architecture artifacts to identify agreements, disagreements, and auto-fixable issues.

What is a dual-model code review and how does it validate architecture artifacts?▼

A dual-model code review independently evaluates architecture artifacts using both Claude and Codex CLI, cross-referencing their outputs to automatically triage clear-cut issues and surface critical disagreements for your final decision.

Do I need Codex CLI installed and authenticated to use automated code triage?▼

Yes, you need Codex CLI installation, authentication, and configuration, along with Claude Code agent teams enabled for parallel review execution to perform automated triage and apply fixes.

Can I automatically fix issues identified by Claude agents and Codex CLI during a review?▼

Yes, the automated review-fix loop identifies auto-fixable issues and applies fixes using the best available skill, presenting only the critical disagreements that require your manual decision.

What is the best way to get a second opinion on code changes before merging a critical feature?▼

The best way to get a second opinion is to use this Skill's simultaneous review feature, which leverages independent quality assessments from Claude and Codex CLI to identify potential bugs or architectural flaws before merging.

When should I avoid using an autonomous review-fix loop for quality assurance?▼

You should avoid using an autonomous review-fix loop when you lack the required Codex CLI configuration or Claude Code agent teams enabled, as the Skill depends on these tools for parallel execution and automated triage.

Related Skills

Explore More Software Engineering Skills→
EwallywEwallyw

omc-teams

Launch multiple CLI workers in parallel tmux panes.

Community
Intermediate
Marshall-HallenbeckMarshall-Hallenbeck
5

review

Analyze uncommitted git diffs for bugs, vulnerabilities, and maintainability issues.

Community
Advanced
heyallencaoheyallencao
118

ks-verify

Run automated test suites and coverage checks to validate code claims.

Community
Intermediate
billydeeii136billydeeii136

codex

Automate code review with AI agents and a pass/fail gate.

Community
Advanced