start

Identify architectural layers and structure of an unfamiliar codebase.

9|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/cliftonc/unwind --skill start-cliftonc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: start
Source: https://github.com/cliftonc/unwind/tree/main/skills/start
Command: npx skills add https://github.com/cliftonc/unwind --skill start-cliftonc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly orient teams when approaching an unfamiliar codebase by identifying its architectural layers, technology choices, and overall structure before deep analysis.

Core Features & Use Cases

  • Architecture discovery: Dispatch a subagent to map layers, entry points, and tech stack to produce a machine-parseable architecture document.
  • Guided onboarding: Establish repository context (remote URL, branch, etc.) to anchor analysis and linking to source.
  • Planning & risk awareness: Output informs layer-by-layer analysis plan and potential migration or refactor tasks.

Quick Start

Dispatch the architecture discovery step to generate docs/unwind/architecture.md.

Frequently Asked Questions about start

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

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

To discover the architecture of an unfamiliar codebase, you can map its structural layers, entry points, and technology choices to generate a machine-parseable architecture document for focused analysis.

What is the best way to start a reverse-engineering project for a new repository?

Starting a reverse-engineering project requires identifying the repository's overall structure and architectural layers first, which establishes the necessary context to anchor deeper layer-by-layer codebase analysis.

Can I generate machine-readable documentation for codebase onboarding and migrations?

Yes, you can generate machine-readable architecture documentation during codebase onboarding to map entry points and tech stack, directly informing your planning for migrations, audits, or modernization tasks.

Does this approach require any specific dependencies or environment setup to map layers?

Mapping architectural layers requires no external dependencies, utilizing an internal subagent to explore the codebase and establish repository context like the remote URL and branch to anchor analysis.

When do I need to identify architectural layers before deep code analysis?

You need to identify architectural layers before deep code analysis when approaching an unfamiliar codebase for onboarding, planning a migration, or conducting an audit to quickly orient your team.