code-review

Generates a custom code review command file for your repository's tech stack and needs.

Updated Mar 4, 2026
One-click install
npx skills add https://github.com/descoped/llm-skills --skill code-review-descoped
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/descoped/llm-skills/tree/main/skills/code-review
Command: npx skills add https://github.com/descoped/llm-skills --skill code-review-descoped

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Generating consistent, project-aware code-review tooling is hard; this skill automates the creation of a Claude Code command tailored to your repository's tech stacks, structure, and architecture, so teams can start rigorous code review workflows with minimal setup.

Core Features & Use Cases

  • Auto-detects repo tech stacks (Rust, Python, Go, JavaScript/TypeScript, React, Next.js, Svelte 5, Swift, Kotlin) and maps them to workspace areas.
  • Produces a single Claude Code command file at .claude/commands/{PREFIX}-code-review.md that orchestrates multi-file analysis across seven universal categories with tech-specific variants.
  • Integrates with existing workflow commands (like /{PREFIX}-issue and /{PREFIX}-start) when present, enabling smooth triage and reporting.
  • Supports iterative regeneration as the repository evolves and structures change.

Quick Start

Use the code-review skill to generate a project-specific /{PREFIX}-code-review Claude Code command for your repository.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I generate a project-specific code review command for my repository?

To generate a project-specific code review command, use this skill to auto-detect your repository's tech stacks, module structure, and architecture, outputting a ready-to-use Claude Code command file.

What tech stacks are supported by the automated code review command generator?

The automated code review command generator supports Rust, Python, Go, JavaScript/TypeScript, React, Next.js, Svelte 5, Swift, and Kotlin, mapping each detected stack to workspace areas.

Does the generated code review command integrate with existing workflow automations?

Yes, the generated code review command integrates with existing workflow automations by referencing commands like issue and start when present, enabling smooth triage and reporting.

How to perform multi-file static analysis across different workspace areas?

Multi-file static analysis is performed by orchestrating the generated command across seven universal categories with tech-specific variants, using dependency ordering based on your architecture.

Can I regenerate the code review command as my repository structure evolves?

Yes, you can regenerate the code review command as your repository evolves and structures change, ensuring the static analysis categories and architecture dependency ordering remain accurate.

What is the output file format for the generated Claude Code static analysis command?

The output file format is a single Markdown file located at .claude/commands/{PREFIX}-code-review.md, containing the full orchestration logic for your repository's deep code quality analysis.