repo-atlas

Generate repo maps, architecture docs, and changelogs under docs/atlas.

32|4|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/cathrynlavery/repo-atlas --skill repo-atlas
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: repo-atlas
Source: https://github.com/cathrynlavery/repo-atlas/tree/main
Command: npx skills add https://github.com/cathrynlavery/repo-atlas --skill repo-atlas

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Build a self-contained persistent context system (atlas) for codebases to help engineers and AI agents understand large repositories quickly with minimal searching.

Core Features & Use Cases

  • Auto-generate a repo map, architecture docs, and changelog for any project.
  • Provide a ready-to-use atlas in docs/atlas for instant contextual onboarding.
  • Integrate agent onboarding by augmenting CLAUDE.md with an atlas workflow.

Quick Start

Map this repository to generate the atlas and onboarding docs.

Frequently Asked Questions about repo-atlas

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

FAQPage Schema
How do I auto-generate architecture docs and a repo map for a large codebase?

Mapping a monorepo creates a persistent contextual atlas that generates architecture docs and onboarding materials, allowing engineers and AI agents to understand large repositories quickly with minimal manual searching.

What is the best way to create onboarding materials for engineers joining a monorepo?

Creating onboarding materials for a monorepo involves generating a persistent atlas in docs/atlas, which provides instant contextual onboarding through auto-generated architecture docs and a 14-day changelog for new engineers.

Can I generate a persistent codebase context map for AI agents without altering runtime behavior?

Generating a persistent codebase context map for AI agents satisfies deterministic workflow phases and operates without altering runtime behavior, safely providing contextual onboarding data for automated codebase understanding.

Do I need external dependencies to map a codebase and generate architecture documentation?

Mapping a codebase and generating architecture documentation requires no external dependencies, utilizing the Python standard library where possible to auto-generate the atlas files for any project size.

How does a persistent atlas integrate agent onboarding into existing codebases?

A persistent atlas integrates agent onboarding by augmenting CLAUDE.md with an atlas workflow, ensuring AI agents receive immediate contextual codebase understanding alongside engineers during the onboarding process.

What specific files are generated when mapping a repository for codebase understanding?

Mapping a repository for codebase understanding auto-generates docs under docs/atlas, specifically creating repo-map.md for architecture context and 08_CHANGELOG_LAST_14_DAYS.md for recent codebase activity tracking.