skills.rest
kelp
kelpCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

cross-review

Coordinate Claude and GPT-5.4 reviews into a merged fix list.

2|1|Updated Jan 30, 2026
Search Tags:#validation#claude#merge#multi-model#diff#code-review#gpt-5.4
One-click install
npx skills add https://github.com/kelp/kelp-claude-plugins --skill cross-review-kelp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-review
Source: https://github.com/kelp/kelp-claude-plugins/tree/main/plugins/cross-review/skills/cross-review
Command: npx skills add https://github.com/kelp/kelp-claude-plugins --skill cross-review-kelp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Multi-model code review with cross-validation. Orchestrates independent Claude and GPT-5.4 reviews, cross-validates findings, and merges them into a single, prioritized fix list. Use when reviewing code changes, auditing files, or wanting a second opinion.

Core Features & Use Cases

  • Independent Claude and GPT-5.4 reviews to surface issues each model uniquely detects
  • Cross-validation and merging findings into a single, actionable fix list for human review
  • Works across diff scopes (uncommitted changes, specific files, or commit ranges) with strict non-editing during evaluation

Quick Start

Provide the code changes you want reviewed and let Cross-Review orchestrate independent model assessments.

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 on my uncommitted changes?▼

Yes, you can cross-validate code diffs by orchestrating independent Claude and GPT-5.4 reviews. The process enforces a strict non-editing policy during evaluation and outputs a normalized, merged fix list with statuses for validation.

Does multi-model code review work with any programming language?▼

Multi-model code review works across code diffs, uncommitted changes, and specific file scopes in any programming language. It normalizes findings from both models into a single, actionable list for human review.

What is the best way to merge findings from Claude and GPT-5.4 during code review?▼

A strict non-editing policy is enforced during the multi-model code review evaluation. This ensures the models only analyze the code diffs and generate a merged fix list without modifying your actual source files.

Why use cross-validation for code review instead of a single model?▼

Cross-validation in code review surfaces high-quality fixes by capturing issues each model uniquely detects. Merging independent Claude and GPT-5.4 reviews provides a broader perspective than relying on a single model's assessment.

Related Skills

Explore More Software Engineering Skills→
iiammaeiiammae

second-opinion

Evaluate code changes with external LLMs like Codex or Gemini.

Community
Advanced
typemdtypemd
9

update-doc

Identify and reconcile discrepancies between implementation code and project documentation.

Official
Intermediate
majiayu000majiayu000
258

git-commit-smart

Generate conventional commit messages from git diff output.

Community
Advanced
MichaelsEngineeringMichaelsEngineering

gpd-validate-conventions

Validate physics convention locks across completed phases for drift.

Community
Advanced