skill-creator

Guide users to create or update Claude skills with structured scaffolds.

2|Updated Jul 30, 2025
One-click install
npx skills add https://github.com/hirokita117/template-ai-coding-config --skill skill-creator-hirokita117
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/hirokita117/template-ai-coding-config/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/hirokita117/template-ai-coding-config --skill skill-creator-hirokita117

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill solves the challenge of creating effective, well-structured AI skills for Claude. It provides a comprehensive guide and best practices, ensuring your custom skills are concise, context-aware, and modular. This streamlines development, reduces errors, and maximizes the efficiency of your AI interactions, ultimately saving you time and effort.

Core Features & Use Cases

  • Structured Skill Design: Guides you through organizing skills with SKILL.md, scripts/, references/, and assets/ for optimal context management and clarity.
  • Progressive Disclosure: Teaches efficient context management by loading information only when needed, preventing token bloat and improving AI performance.
  • Workflow & Output Patterns: Provides proven patterns for designing sequential/conditional workflows and ensuring consistent, high-quality AI-generated outputs.
  • Use Case: A developer needs to create a new Claude skill to automate interactions with a proprietary internal API. This skill guides them step-by-step, from defining the API schema in references/ and writing helper functions in scripts/, to structuring the SKILL.md to ensure Claude uses the API effectively and securely.

Quick Start

I need to create a new Claude skill. Guide me through the process of defining its purpose, structure, and content.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I structure a Claude skill for effective context management?

Structure your Claude skill with a SKILL.md file, scripts/ folder for helper functions, references/ folder for domain schemas or APIs, and assets/ folder for supporting files. This modular organization prevents token bloat and ensures Claude loads information progressively only when needed.

What are the required fields and naming conventions for creating a Claude skill?

Claude skills require a name field (hyphen-case format) and a description with specific length limits, validated through frontmatter checks. These requirements ensure consistency and compatibility with skill validation and packaging scripts.

Can I use Claude skills to automate interactions with proprietary APIs?

Yes. Define your API schema in the references/ folder, write helper functions in scripts/, and structure your SKILL.md to guide Claude on secure and effective API usage. This approach ensures Claude understands your API's requirements and constraints.

How do I design workflows and output patterns for consistent AI-generated results?

Use proven sequential and conditional workflow patterns provided by the skill development guide. Structure tasks clearly in your SKILL.md and leverage scripts/ for reusable logic to ensure Claude produces consistent, high-quality outputs across repeated interactions.

What dependencies do I need to create and validate a Claude skill?

You need PyYAML to handle frontmatter validation and skill configuration. The skill includes quick_validate and packaging scripts that work with this dependency to bundle your resources and validate your skill structure before deployment.

How do I prevent token bloat when designing AI skills?

Apply progressive disclosure by organizing information across SKILL.md, scripts/, and references/ folders. Load detailed content only when Claude needs it, reducing unnecessary context and improving AI performance and response latency.