If you've ever onboarded onto a codebase and wished someone had written down what each directory actually does, this automates that across your entire project. It generates AGENTS.md files at every level with file descriptions, dependency maps, and AI-specific instructions for working in each area. The hierarchical parent tagging system means agents can navigate up and down the tree. Smart about it too: skips empty directories, preserves your manual notes when regenerating, and processes directories level by level so parent references stay valid. The validation checks catch orphaned files and broken links. Basically turns your file tree into a documented, navigable knowledge graph that both humans and AI agents can actually use.

skills.sh installs
1.1K
GitHub stars
40.7K
last updated
Aug 21, 2026

Install

$ npx -y skills add yeachan-heo/oh-my-claudecode --skill deepinit --agent claude-code

Installs into .claude/skills of the current project.

Files

Deep Init Skill

Creates comprehensive, hierarchical AGENTS.md documentation across the entire codebase.

Core Concept

AGENTS.md files serve as AI-readable documentation that helps agents understand:

  • What each directory contains
  • How components relate to each other
  • Special instructions for working in that area
  • Dependencies and relationships

Hierarchical Tagging System

Every AGENTS.md (except root) includes a parent reference line:

**Parent context:** `../AGENTS.md`

This MUST be visible prose, never an HTML comment. Claude Code strips HTML comments before a memory file reaches the model, so <!-- Parent: ... -->, <!-- Generated: ... --> and a commented MANUAL boundary are invisible to the agent that is supposed to act on them.

This creates a navigable hierarchy:

/AGENTS.md                          ← Root (no parent line)
├── src/AGENTS.md                   ← **Parent context:** `../AGENTS.md`
│   ├── src/components/AGENTS.md    ← **Parent context:** `../AGENTS.md`
│   └── src/utils/AGENTS.md         ← **Parent context:** `../AGENTS.md`
└── docs/AGENTS.md                  ← **Parent context:** `../AGENTS.md`

Loading Model

Sponsors

Details

Repository
github.com/yeachan-heo/oh-my-claudecode
Category
Documentation
First seen
Jun 3, 2026

More from yeachan-heo/oh-my-claudecode

All 37
1cap-go avatarcapacitor-keyboardcap-go/capgo-skillsGuide to handling keyboard in Capacitor apps including visibility detection, accessory bar, scroll behavior, and input focus. Use this skill when users have key1.1Kskills.sh installs
2daymade avatarcapture-screendaymade/claude-code-skillsProgrammatic screenshot capture on macOS. Find window IDs with Swift CGWindowListCopyWindowInfo, control application windows via AppleScript (zoom, scroll, sele1.1Kskills.sh installs
3rstackjs avatarrslib-best-practicesrstackjs/agent-skillsRslib best practices for config, CLI workflow, output, declaration files, dependency handling, build optimization and toolchain integration. Use when writing, r1.1Kskills.sh installs
4asyrafhussin avatarlaravel-owasp-securityasyrafhussin/agent-skillsOWASP Top 10 security audit and secure coding guidelines for Laravel + React/Inertia.js applications. Use when auditing for vulnerabilities ("run OWASP audit", 1.1Kskills.sh installs
5inference-sh avatarai-marketing-videosinference-sh/skillsCreate AI marketing videos for ads, promos, product launches, and brand content. Models: Veo, Seedance, Wan, FLUX for visuals, Kokoro for voiceover. Types: prod1.1Kskills.sh installs
6inference-sh avatarprompt-engineeringinference-sh/skillsMaster prompt engineering for AI models: LLMs, image generators, video models. Techniques: chain-of-thought, few-shot, system prompts, negative prompts. Models:1.1Kskills.sh installs