How should I name my new Claude skill?

Follow the lowercase-hyphen convention (e.g., 'processing-pdfs' or 'github-pr-creation'). Use 2-4 words that are descriptive but concise.

What is the token budget for a skill file?

Ideally, the SKILL.md file should be under 500 lines. Detailed documentation or examples should be moved to a separate 'references' directory to save context window space.

When should I use 'context: fork' in a skill?

Use 'context: fork' for skills that process large amounts of data, such as PDFs or images, to prevent the main conversation context from becoming bloated.

Why is the description field in the frontmatter so important?

The description is the primary mechanism for discovery. It must include what the skill does and specific trigger phrases that help Claude decide when to activate it.

What is a Claude Code Skill?

A Claude Code Skill is a specialized capability that enhances Claude's behavior for specific tasks by providing domain-specific guidance, best practices, and implementation patterns.

Claude Skill Creator

Name: Claude Skill Creator
Author: Bae-ChangHyun

byBae-ChangHyun

•

Developer Tools

Streamlines the development of custom Claude Code skills by providing standardized templates, naming conventions, and official best practices.

The Claude Skill Creator is a comprehensive resource designed to help developers build high-quality, standardized skills for the Claude Code CLI. It provides a structured framework for creating SKILL.md files, defining effective trigger phrases for discovery, and optimizing token usage through progressive disclosure. By following Anthropic's official guidelines, this skill ensures that custom extensions are efficient, maintainable, and follow consistent naming and directory conventions, making it an essential tool for developers looking to extend Claude's capabilities in their local workflow.

Key Features

01Support for advanced v2.1.0 features including context forking and bash wildcard permissions

021 GitHub stars

03Best practices for Python helper scripts with structured JSON output requirements

04Standardized SKILL.md template with required frontmatter and structured body sections

05Token budget management guidelines to keep skills efficient and under 500 lines

06Optimized discovery logic using specific trigger phrase formulas and naming conventions

Use Cases

01Refactoring complex skills to improve reliability and reduce token consumption

02Bootstrapping a new custom capability for Claude Code from scratch

03Standardizing personal automation scripts into formal, discoverable Claude skills

Key Features

01Support for advanced v2.1.0 features including context forking and bash wildcard permissions

021 GitHub stars

03Best practices for Python helper scripts with structured JSON output requirements

04Standardized SKILL.md template with required frontmatter and structured body sections

05Token budget management guidelines to keep skills efficient and under 500 lines

06Optimized discovery logic using specific trigger phrase formulas and naming conventions

Use Cases

01Refactoring complex skills to improve reliability and reduce token consumption

02Bootstrapping a new custom capability for Claude Code from scratch

03Standardizing personal automation scripts into formal, discoverable Claude skills