What is a Claude Code Skill?

A Claude Code Skill is a modular, self-contained package that extends Claude's capabilities by providing specialized knowledge, workflows, and tool integrations for specific domains.

When should I include scripts in a skill?

Include scripts when a task requires deterministic reliability, or when Claude would otherwise have to repeatedly rewrite the same complex code for a specific operation.

What is the Progressive Disclosure principle in skills?

It is a design pattern that manages context efficiently by only loading the most relevant information at each stage: starting with metadata, then the SKILL.md body, and finally specific resources as needed.

How do I keep my skills token-efficient?

Keep the main SKILL.md file concise (under 500 lines) and use reference files for detailed documentation, schemas, and examples that Claude can load only when necessary.

What files are required for a Claude Skill?

Every skill requires a SKILL.md file with YAML frontmatter containing a name and description. Optional directories include scripts, references, and assets.

Claude Skill Creation Guide

Name: Claude Skill Creation Guide
Author: notedit

bynotedit

•

229

学習とドキュメンテーション

Provides a standardized framework and best practices for building modular, high-performance specialized capabilities for Claude Code.

概要

This skill serves as the definitive guide for developers looking to extend Claude's capabilities with specialized knowledge, workflows, and tool integrations. It outlines the core principles of skill architecture, including the Progressive Disclosure design pattern which optimizes context window usage. By following this guide, users can transform Claude from a general-purpose assistant into a domain-specific agent equipped with procedural knowledge, bundled scripts, and organized reference materials for complex coding and automation tasks.

主な機能

Comprehensive blueprints for skill anatomy including SKILL.md, scripts, and assets
Guidance on setting appropriate degrees of freedom for agent behavior
Instructional patterns for organizing domain-specific documentation and schemas
Standardized workflows for initializing, editing, and packaging new skills
229 GitHub stars
Best practices for context window management and token efficiency

ユースケース

Creating reusable toolsets for complex data transformations or file processing
Optimizing existing skills to improve reliability and reduce context bloat
Designing custom agents for proprietary company APIs or internal workflows

概要

主な機能

Comprehensive blueprints for skill anatomy including SKILL.md, scripts, and assets
Guidance on setting appropriate degrees of freedom for agent behavior
Instructional patterns for organizing domain-specific documentation and schemas
Standardized workflows for initializing, editing, and packaging new skills
229 GitHub stars
Best practices for context window management and token efficiency

ユースケース

Creating reusable toolsets for complex data transformations or file processing
Optimizing existing skills to improve reliability and reduce context bloat
Designing custom agents for proprietary company APIs or internal workflows