---

🌟 Overview

The skills repository is designed around progressive disclosure.
Each skill retains core operational instructions in SKILL.md while deep-dive context, scripts, and templates are structured into dedicated references/, scripts/, or assets/ directories. This prevents context bloat, reduces LLM hallucination, and ensures high-quality generation.

---

🧭 Repository Structure

📦 skills
 ┣ 📂 .agents/              # Local agent configuration files
 ┣ 📂 .devcontainer/        # Development environment configuration
 ┣ 📂 .github/              # GitHub-specific configuration
 ┃  ┗ 📂 agents/            # Reusable agent definition files (.agent.md)
 ┣ 📂 skills/               # Core agent skills
 │  ┣ 📂 skill-crafter/
 │  ┣ 📂 agents-md-crafter/
 │  ┣ 📂 readme-crafter/
...

---

🛠️ Available Skills

Domain & Skill	Description
🤖 Agent Engineering
🧭 programming-guidelines	Universal coding standards to apply before editing, refactoring, debugging, or generating code.
🪛 skill-crafter	Create, edit, evaluate, and optimize new Agent Skills efficiently.
📜 agents-md-crafter	Design and update standard AI agent instruction files like AGENTS.md.
📝 readme-crafter	Focused documentation agent for creating and revising README.md files.
✅ conventional-commits	Standardize git commit messages with Conventional Commits 1.0.0 and 50/72 rules.
🧰 Dev Environment
📦 devcontainer-creator	Create, edit, and update .devcontainer/devcontainer.json development environments.
🐍 Python Ecosystem
📜 python-docstring-expert	Expert methodology for evaluating, formatting, and generating Python docstrings.
💬 python-comment-expert	Expertise in writing, reviewing, and modifying Python logic code comments.
🇬 python-google-docstring-crafter	Google Style Python Docstring expertise: modules, classes, functions, and formatting.
⚙️ Security
🛡️ data-leakage-prevention	Audit staged changes, commits, directories, or repos for secrets, PII, binary-document exposure, and git identity leakage.

---

⚡ Quick Start

1️⃣ Development Container (Recommended)

This repository includes a robust Dev Container configured with Ubuntu, Python, Pre-commit, Git LFS, and AI agent dependencies.

1. Open this repository in Visual Studio Code.
2. Hit Ctrl+Shift+P (or Cmd+Shift+P on Mac).
3. Search and select: Dev Containers: Reopen in Container.

2️⃣ Local Setup

If developing locally:

# Install pre-commit globally or in a virtualenv
pip install pre-commit

# Install git hook scripts
pre-commit install

# Run against all files manually
pre-commit run --all-files

---

✍️ Creating a New Skill

Follow our standard directory topography when authoring new instructions:

1. SKILL.md: Keep it minimal and actionable. This defines the agent's persona, strict rules, and routing logic.
2. references/: Heavy documentation and rulesets. Read dynamically by agents only when needed.
3. assets/: Executable templates, JSON schemas, or example configurations.
4. scripts/: Helpers (e.g., Python validation scripts).

💡 Tip: Refer to the skill-crafter to help generate your new skills!

---

Managed by AI, built for AI.
Check out AGENTS.md to understand how AI operates within this repository.

For CI, repository workflows also run secret scanning checks.

🤝 Contributing

Contributions are welcome! Suggested flow:

1. Create or update a skill under skills/
2. Keep changes focused and minimal
3. Run pre-commit run --all-files
4. Open a pull request with a clear summary and rationale

If you are creating a new skill, use the custom agent definition in .github/agents/agent-instructions-maintainer.agent.md as a structural reference.

📄 License

Licensed under the Apache License 2.0. See LICENSE for details.