review

Run parallel code reviews with Claude, Gemini CLI, and Codex CLI.

Updated Dec 17, 2025
One-click install
npx skills add https://github.com/eprouveze/HealthPulse --skill review-eprouveze
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/eprouveze/HealthPulse/tree/main/.claude/skills/review
Command: npx skills add https://github.com/eprouveze/HealthPulse --skill review-eprouveze

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill performs multi-model code review by running Claude, Gemini CLI, and Codex CLI in parallel to identify issues across code changes, then merges and deduplicates findings to accelerate high-quality releases.

Core Features & Use Cases

  • Parallel analysis across Claude, Gemini CLI, and Codex CLI with automatic merging and de-duplication of findings
  • Priority triage into P1/P2/P3 with disagreement highlighting
  • Fallback to Claude-only when external CLIs are unavailable
  • Use after implementing a feature, before committing, or when asked to review

Quick Start

Activate on your uncommitted changes to trigger a parallel, multi-model review and return triaged findings.

Frequently Asked Questions about 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?

Multi-model code review executes Claude, Gemini CLI, and Codex CLI in parallel on your uncommitted changes, then merges and deduplicates findings into a triaged report with P1/P2/P3 priorities.

What is priority triage in a multi-model code review?

Priority triage categorizes consolidated code review findings into P1, P2, and P3 severity levels, highlighting disagreements between Claude, Gemini CLI, and Codex CLI.

Do I need Gemini CLI and Codex CLI installed to use this multi-model review?

You do not need Gemini CLI and Codex CLI installed; the review workflow automatically falls back to Claude-only analysis when those external CLIs are unavailable.

Can I apply project-specific brand rules during the code review?

Yes, the multi-model code review respects project-specific rules defined in the .claude/brand.json file when it is present in your project directory.

When is the best time to run a parallel multi-model code review?

The best time to run a parallel multi-model code review is after implementing a feature, before committing, or when explicitly asked to review code changes pre-merge.