skill-creator

Create, organize, and package modular Claude skills with Python scripts.

215|34|Updated Jan 26, 2026
One-click install
npx skills add https://github.com/sjtu-sai-agents/EvoMaster --skill skill-creator-sjtu-sai-agents
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/sjtu-sai-agents/EvoMaster/tree/main/evomaster/skills/skill-creator
Command: npx skills add https://github.com/sjtu-sai-agents/EvoMaster --skill skill-creator-sjtu-sai-agents

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill solves the challenge of building, structuring, and packaging self-contained Claude skills to extend capabilities in a modular, reusable way.

Core Features & Use Cases

  • Provides a standard SKILL.md frontmatter framework and a structured body for instructive guidance.
  • Includes optional bundles (scripts/, references/, assets/) to support repeatable workflows and scalable distribution.
  • Offers a complete lifecycle: initialization with init_skill.py, validation, and packaging into .skill files for distribution.

Quick Start

Run the skill initializer to create a new skill scaffold, then customize SKILL.md and resources. Example: python scripts/init_skill.py my-new-skill --path skills/public. Then verify structure with the validator and package as needed using the provided packaging script.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create and package modular Claude skills for distribution?

To create and package modular Claude skills, use an initialization script to scaffold directories, customize the SKILL.md frontmatter, validate the structure, and package the output into distributable .skill files.

What is the standard frontmatter framework needed for skill creation?

The standard frontmatter framework for skill creation requires structured metadata fields enforced by a validator to ensure modular extensions are correctly organized before they can be packaged and distributed.

Can I include optional scripts and assets when building modular extensions?

Yes, you can include optional scripts, references, and assets directories when building modular extensions to support repeatable workflows and scalable distribution alongside the core SKILL.md file.

Do I need Python to initialize and validate a new skill scaffold?

Yes, you need Python to execute the provided init_skill.py script for generating a new skill scaffold and to run the validation scripts that verify the standardized directory structure before packaging.

What's the best way to organize self-contained Claude skills for repeatable workflows?

The best way to organize self-contained Claude skills is using a structured lifecycle approach: initialize a scaffold, populate the SKILL.md body with instructive guidance, add optional resource bundles, and validate before packaging.

Why does skill validation fail when packaging modular extensions?

Skill validation fails during packaging when the SKILL.md frontmatter metadata does not match the standardized framework or when the directory structure lacks required components enforced by the validation scripts.