describe-codebase

Explain codebase structure, data storage, and conventions across three lenses.

2|Updated Jun 5, 2026
One-click install
npx skills add https://github.com/cleanunicorn/agents-library --skill describe-codebase
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: describe-codebase
Source: https://github.com/cleanunicorn/agents-library/tree/main/skills/describe-codebase
Command: npx skills add https://github.com/cleanunicorn/agents-library --skill describe-codebase

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides a deep understanding of a codebase's structure, layers, data storage, and conventions, allowing users to quickly grasp the overall design and architecture.

Core Features & Use Cases

  • Comprehensive Overview: Maps the entire codebase or a specific subsystem across three lenses (layering, data, conventions).
  • Scalable Scopes: Supports whole repository, path/glob for a subsystem, or tracing a single execution path.
  • Read-only: Never modifies code and writes documentation only on explicit confirmation.
  • Use Case: Ideal for onboarding new team members, understanding complex subsystems, or writing an architecture document.

Quick Start

Run the 'describe-codebase' skill on the entire repository or a specific path: describe-codebase /path/to/subsystem.

Frequently Asked Questions about describe-codebase

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

FAQPage Schema
How do I quickly understand the architecture of an unfamiliar codebase?

You can analyze the codebase structure across three lenses: layering, data storage, and conventions. This maps the entire repository or specific subsystem to provide a consolidated brief of the overall design.

Can I trace a single execution path through a subsystem without modifying the code?

Yes, you can trace a single execution path without modifying the code. The codebase analysis is strictly read-only and generates documentation only when explicitly confirmed.

What is the best way to generate architecture documentation for onboarding new team members?

The best way to generate architecture documentation for onboarding is to map the codebase structure, layers, and conventions, producing an optional consolidated brief or documentation file.

Does codebase analysis support mapping specific directories or glob patterns instead of the whole repository?

Yes, codebase analysis supports mapping specific directories or glob patterns instead of the whole repository, allowing targeted subsystem mapping or execution path tracing.

How do I detect coding conventions and layering rules in an existing system?

You detect coding conventions and layering rules by applying a three-lens analysis covering layering, data storage, and conventions across your specified project scope.