review

Aggregate persona-driven reviewer perspectives into prioritized code critiques with citations.

Updated Nov 13, 2025

Search Tags:#automation #analysis #claude #perspectives #best-practices #code-review #reference-load

Ready to run

npx skills add https://github.com/zacjones93/pokidex --skill review-zacjones93

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review
Source: https://github.com/zacjones93/pokidex/tree/main/.claude/skills/review
Command: npx skills add https://github.com/zacjones93/pokidex --skill review-zacjones93

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill centralizes persona-driven code reviews (Carmack, React core, Anders, React Router, etc.) so Claude can select or combine expert viewpoints when the user asks for a code review or perspective-specific critique.

Core Features & Use Cases

Unified reviewer personas: Claude activates this Skill and lazy-loads the exact perspective by opening the reference docs linked below or by spawning persona-specific subagents.
Context-aware critique: Claude bases evaluations on the loaded perspective and applies the priorities from the persona's reference.
Multi-perspective synthesis: When multiple personas are requested, Claude synthesizes them into a single, actionable critique with citations.

Quick Start

Ask Claude to perform a code review. Claude will first prompt you to load the relevant persona reference file(s) using the Read tool, then apply the chosen perspectives to the code and deliver a multi-perspective critique with concrete recommendations and citations.

Frequently Asked Questions about review

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

FAQPage Schema

How do I get a code review from multiple expert perspectives?▼

Code review can aggregate perspectives from personas like Carmack, React core, and Anders Hejlsberg by loading their reference files. Claude applies each perspective's priorities—accessibility, type-safety, testing, performance, UI/router concerns—then synthesizes them into a single actionable critique with citations.

What makes persona-driven code review different from standard review?▼

Persona-driven code review loads specific expert reference files that encode each reviewer's priorities and heuristics. This ensures bias-resistant, comprehensive critique grounded in concrete guidance rather than generic feedback, with recommendations tied directly to each persona's standards.

How do I start a code review with this Skill?▼

Ask Claude to perform a code review and specify which personas you want (Carmack, React core, etc.). Claude loads the relevant reference files, parses reviewer hints, then delivers a structured critique with concrete recommendations and line citations.

Can I combine multiple reviewer perspectives in one review?▼

Yes, code review synthesizes multiple personas into a single prioritized critique. When you request multiple perspectives, Claude merges their guidance, resolves conflicts, and produces unified recommendations with citations to each source.

What code concerns does this Skill evaluate?▼

Code review applies persona guidance across accessibility, type-safety, testing, performance, and UI/router concerns. Each loaded perspective prioritizes different aspects, so the final critique reflects the combined expertise of all activated personas.

Do I need reference files to use this Skill?▼

Yes, code review requires loading persona reference files via the Read tool. These files contain each reviewer's guidance; Claude lazy-loads the exact perspectives you request, so only relevant reference docs are opened.