What is LLM-first design in skill creation?

It is a documentation strategy that prioritizes high information density, structural patterns, and the removal of pedagogical fluff to save tokens and improve AI comprehension.

How do I initialize a new skill using this toolkit?

You can use the included 'init_skill.py' script to generate a standardized directory structure including SKILL.md, scripts, and references.

When should I use a script versus a reference file?

Use scripts for deterministic logic or repeated code execution; use references for documentation, schemas, or policies that Claude needs to consult.

Why is YAML frontmatter required in SKILL.md?

Frontmatter contains the name and description used by Claude to programmatically discover and trigger the skill based on specific keywords.

What is a Claude Code Skill?

A Claude Code Skill is a specialized capability that enhances Claude's behavior using domain-specific guidance, scripts, and implementation patterns.

Skill Creation & Quality Guide

Name: Skill Creation & Quality Guide
Author: Mission42-ai

byMission42-ai

개발자 도구

Provides a comprehensive framework and automated tooling for building high-density, LLM-optimized Claude Code skills.

소개

The Skill Creation plugin is a meta-toolkit designed for developers building modular Claude Code workflows. It enforces strict metadata standards, promotes LLM-first documentation patterns to maximize token efficiency, and provides a structured 5-step process for initializing, developing, and reviewing new agent capabilities. By focusing on progressive disclosure and information density, it ensures that custom skills are both discoverable and highly effective when executed by AI agents.

주요 기능

Multi-level progressive disclosure loading logic
LLM-optimized documentation patterns for maximum information density
Standardized resource management for scripts, assets, and references
Automated skill initialization via CLI scaffolding scripts
Strict metadata validation for improved skill discoverability
0 GitHub stars

사용 사례

Optimizing existing documentation for better AI agent performance
Scaffolding new Claude Code plugins with standardized directory structures
Conducting quality reviews on skill metadata and YAML frontmatter

소개

주요 기능

Multi-level progressive disclosure loading logic
LLM-optimized documentation patterns for maximum information density
Standardized resource management for scripts, assets, and references
Automated skill initialization via CLI scaffolding scripts
Strict metadata validation for improved skill discoverability
0 GitHub stars

사용 사례

Optimizing existing documentation for better AI agent performance
Scaffolding new Claude Code plugins with standardized directory structures
Conducting quality reviews on skill metadata and YAML frontmatter