learn

Coordinate multiple Haiku agents to explore a codebase and generate Markdown documentation.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/heman1033p-create/heman1033p-create.github.io --skill learn-heman1033p-create
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn
Source: https://github.com/heman1033p-create/heman1033p-create.github.io/tree/main/.gemini/skills/learn
Command: npx skills add https://github.com/heman1033p-create/heman1033p-create.github.io --skill learn-heman1033p-create

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Exploring large codebases manually is time-consuming and error-prone; this Skill coordinates parallel Haiku agents to automatically generate organized documentation, speeding up orientation and maintenance.

Core Features & Use Cases

  • Parallel Haiku agents (1/3/5) for fast, default, and deep explorations.
  • Auto-source acquisition via ghq and origin/symlinks to keep repos in sync.
  • Time-stamped, date-foldered documentation hub with per-run outputs.

Quick Start

Use a repository URL or slug to start automated exploration of the codebase.

Frequently Asked Questions about learn

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

FAQPage Schema
How do I automate codebase exploration to generate documentation?

You can automate codebase exploration by running 1, 3, or 5 parallel Haiku agents that analyze a repository and output structured Markdown documentation. This replaces manual code review with coordinated multi-agent processing.

What is the best way to onboard a team to a large new codebase?

Onboarding to a large codebase is best handled by running deep multi-agent exploration to generate a time-stamped documentation hub. This coordinates 5 parallel agents to map the repository architecture into organized Markdown outputs.

Does this codebase documentation tool support different exploration depths?

Yes, this codebase documentation tool supports --fast, default, and --deep modes. These coordinate 1, 3, or 5 parallel Haiku agents respectively, letting you balance exploration speed against documentation thoroughness.

How do I start exploring a repository using a URL or slug?

To start exploring a repository, provide a repository URL or slug. The tool uses ghq to automatically clone or link the sources via symlinks, ensuring the codebase is synced before initiating the multi-agent documentation workflow.

How are the generated codebase documentation outputs organized?

Generated codebase documentation outputs are organized into a time-stamped date folder with a central hub index. This structure stores per-run Markdown files chronologically, making it easy to track documentation history across explorations.