skill-creator

Generate Claude Skill structures with YAML frontmatter and validation.

7|3|Updated Oct 23, 2025
One-click install
npx skills add https://github.com/BerryKuipers/claude-code-toolkit --skill skill-creator-berrykuipers
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/BerryKuipers/claude-code-toolkit/tree/main/.claude/skills/meta/skill-creator
Command: npx skills add https://github.com/BerryKuipers/claude-code-toolkit --skill skill-creator-berrykuipers

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires grep, jq, sed.

What problem does it solve?

Manually creating new Claude Skills with proper structure, YAML frontmatter, progressive disclosure, and validation can be complex and error-prone. This interactive wizard streamlines the entire skill creation process.

Core Features & Use Cases

  • Interactive Wizard: Guides users through skill creation by asking clarifying questions about the workflow to be automated.
  • Automated Structure Generation: Validates skill name and description, determines category and complexity, and generates the correct directory structure and SKILL.md template.
  • Progressive Disclosure: Optionally creates REFERENCE.md for complex skills and validation scripts in scripts/ to reduce token consumption.
  • Use Case: When you identify a repetitive task or a common workflow pattern, use this Skill to quickly bootstrap a new, well-structured Claude Skill, ready for integration into agents and workflows.

Quick Start

Start the interactive skill creation wizard

skill-creator

The skill will prompt you with questions like:

1. What workflow or task do you want to automate?

2. When should this skill be triggered?

3. What tools/commands does this workflow use?

...

Based on your answers, it will generate the skill files in the correct structure.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I automate Claude Skill creation with proper structure and validation?

Skill creation automation validates your skill name and description, assigns category and complexity levels, and generates a complete directory structure with YAML frontmatter and templates. The interactive wizard guides you through workflow requirements to produce well-structured, reusable skills ready for integration.

What does progressive disclosure mean in skill templates?

Progressive disclosure generates optional reference documentation and validation scripts separately from core skill files, reducing token consumption by keeping detailed content out of the main skill logic. Complex skills receive a REFERENCE.md file and scripts/ validators to offload non-essential processing.

Can I use this to convert existing workflows into Claude Skills?

Yes. The interactive wizard interviews you about your existing workflow—what it automates, when it triggers, which tools it uses—then standardizes your process into a properly structured Claude Skill with YAML frontmatter and validation, eliminating manual structure work.

What tools and dependencies does the skill creation process require?

Skill creation uses grep, jq, and sed for template generation, validation, and YAML processing. These command-line utilities handle naming validation, category assignment, and structure generation without additional external dependencies.

How does this help standardize skill creation across teams?

The wizard enforces consistent naming conventions, category assignment, complexity estimation, and YAML frontmatter formatting. All generated skills follow the same template structure, ensuring team-wide standardization and reducing review friction during integration.

What happens if my skill requires complex logic or multiple references?

For intermediate and complex skills, the wizard optionally generates a separate REFERENCE.md file documenting intricate workflows and a scripts/ directory with validation logic. This structure keeps the main skill definition lean while preserving detailed reasoning separately.