skill-creator

Create and update Claude skills with SKILL.md and optional resources.

6|Updated Nov 27, 2025
One-click install
npx skills add https://github.com/Holo00/IdeaForge --skill skill-creator-holo00
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Holo00/IdeaForge/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/Holo00/IdeaForge --skill skill-creator-holo00

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill guides users through the process of creating effective and token-efficient Claude Skills, extending AI capabilities with specialized knowledge, workflows, and tool integrations.

Core Features & Use Cases

  • Skill Anatomy: Understand the modular structure of a Skill, including SKILL.md and optional resources (scripts/, references/, assets/).
  • Core Principles: Learn to write concise, context-aware skills and set appropriate degrees of freedom for optimal AI performance.
  • Progressive Disclosure: Implement patterns to manage context window usage effectively, ensuring Claude only loads necessary information.
  • Use Case: "Develop a new skill for managing project tasks, including a SKILL.md that outlines the workflow and a scripts/ directory for common task automation, making Claude a more effective project assistant."

Quick Start

Use the skill-creator skill to outline a new skill named 'code-reviewer' that helps Claude review Python code for best practices.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create a Claude Skill with YAML frontmatter and Markdown guidance?

A Claude Skill is a structured package combining a frontmatter YAML header, a SKILL.md file with imperative instructions, and optional resource directories (scripts/, references/, assets/). Use skill-creator to automate this structure, ensuring your SKILL.md stays concise while loading references on demand to manage context window usage efficiently.

What's the best way to organize a Skill's folder structure for tool integrations and workflows?

Organize your Skill with a SKILL.md body for core guidance, a scripts/ directory for automation procedures, a references/ directory for knowledge resources, and an assets/ directory for supporting files. This modular structure lets Claude load only necessary information, keeping tokens lean while supporting multi-step procedures and API interactions.

How do I write token-efficient Claude Skills that manage context effectively?

Apply progressive disclosure: write concise SKILL.md guidance, set appropriate degrees of freedom for Claude's reasoning, and structure resources so Claude loads references on demand rather than upfront. skill-creator enforces these best practices, helping you extend Claude's capabilities without bloating the context window.

Can I use skill-creator to update an existing Skill or only create new ones?

skill-creator automates both creation and updating of Claude Skills. Whether you're building a new Skill from scratch or refining an existing one to add workflows, tool integrations, or domain knowledge, the Skill enforces the same frontmatter YAML and Markdown structure while maintaining best practices.

Do I need programming experience to create a Skill with this approach?

No. skill-creator guides you through Skill anatomy and core principles without requiring advanced coding. The focus is on writing clear imperative guidance in Markdown, organizing resources logically, and understanding how to set degrees of freedom—skills applicable across specialized workflows, data processing, and template generation.

What's the difference between storing guidance in SKILL.md versus bundled resource files?

SKILL.md contains the core imperative guidance Claude reads directly, kept lean for efficiency. Bundled resources in scripts/, references/, and assets/ directories hold procedures, reference data, and supporting materials that Claude loads on demand. This separation prevents context bloat while keeping specialized knowledge accessible for complex workflows and API interactions.