code-review

Run multi-layer code reviews with independent subagents per layer.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/alexlazarian/claude-skills --skill code-review-alexlazarian
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/alexlazarian/claude-skills/tree/main/code-review
Command: npx skills add https://github.com/alexlazarian/claude-skills --skill code-review-alexlazarian

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Entropy-reducing code review orchestrates architecture-aware, multi-layer evaluations that favor deletion, consolidation, and simplification over additive fixes.

Core Features & Use Cases

  • Independent subagents per layer (architecture, data flow, testability, security, correctness, test coverage, performance, observability, and code hygiene) operate with fresh context and isolated findings.
  • Sequential orchestration with a transparent triage policy that decides which layers to run and in what order.
  • Actionable guidance that surfaces targeted improvements to reduce system entropy and streamline codebases.
  • Use Case: when refactoring or introducing changes, the skill surfaces cross-cutting impacts and suggests consolidation opportunities to minimize complexity.

Quick Start

Summarize the latest changes, run layer-wise triage to select applicable reviews, and execute the subagent-based assessments in sequence.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I run a multi-layer code review on recent git changes?

To run a multi-layer code review, the skill analyzes recent changes with git diff, selects relevant review layers via triage, and executes independent subagents in sequence to report findings.

What is entropy-reducing code review and when do I need it?

Entropy-reducing code review is an architecture-aware evaluation that favors deletion and consolidation over additive fixes. You need it when refactoring to surface cross-cutting impacts and minimize complexity.

How does architecture-aware code review handle different evaluation layers?

Architecture-aware code review coordinates independent subagents for each layer, such as data flow, security, and performance. It operates with fresh context and isolated findings to provide targeted remediation guidance.

Can I use this code review skill for quality assurance and refactoring?

Yes, you can use this skill for quality assurance and refactoring. It orchestrates multi-layer evaluations that surface consolidation opportunities and suggest targeted improvements to reduce system entropy.

What is the best way to simplify code architecture during a review?

The best way to simplify code architecture during a review is using a transparent triage policy that selects applicable review layers, executing subagent-based assessments sequentially to streamline the codebase.

Does this code review approach support testability and observability analysis?

Yes, this code review approach supports testability and observability analysis by running independent subagents with fresh context for each layer, ensuring isolated findings and actionable guidance.