skill-builder

Create and refine Claude Code skills with SKILL.md definitions.

1|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/ztdevelops/claude-code-skills --skill skill-builder-ztdevelops
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-builder
Source: https://github.com/ztdevelops/claude-code-skills/tree/main/public/skill-builder
Command: npx skills add https://github.com/ztdevelops/claude-code-skills --skill skill-builder-ztdevelops

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Helps authors and teams consistently create, convert, and maintain Claude Code Skill directories so Claude reliably discovers and invokes the right capability without confusion or duplication.

Core Features & Use Cases

  • Skill authoring guidance: Step-by-step assistance to design skill workflows and write valid SKILL.md frontmatter that uses gerund naming and invocation-focused descriptions.
  • Skill editing and validation: Techniques to refine descriptions, apply progressive disclosure, and split long content into intention-revealing reference files.
  • Sub-agent conversion: Concrete transformation steps for converting sub-agents into skills, removing model/tools fields, and emphasizing Node.js and CLI tooling.
  • Real-world scenario: Create a new global skill, shorten an overlong SKILL.md into pointers + reference files, or convert a local sub-agent into a well-structured skill for project-wide use.

Quick Start

Use this skill to draft a new SKILL.md, pick a gerund-form name, craft a third-person invocation description starting with "Use this skill when...", and outline any supporting reference files and Node.js scripts.

Frequently Asked Questions about skill-builder

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

FAQPage Schema
How do I create a Claude Code skill with a valid SKILL.md file?

To create a Claude Code skill, you draft a SKILL.md file using gerund-form naming and an invocation-focused description starting with "Use this skill when...". This ensures Claude reliably discovers and invokes the capability without confusion.

What is progressive disclosure when authoring skills for CLI workflows?

Progressive disclosure is the practice of keeping SKILL.md concise and splitting long content into intention-revealing reference files. This structure ensures reliable automatic invocation by prioritizing Node.js-centric scripting and CLI patterns for deterministic tasks.

Can I convert a Claude Code sub-agent into a Node.js skill?

Yes, you can convert a sub-agent into a skill by removing model and tools fields, then restructuring the content into a SKILL.md with supporting reference files. This emphasizes Node.js and CLI tooling for deterministic, project-wide workflows.

How do I fix a Claude Code skill that fails to trigger automatically?

To fix an unresponsive Claude Code skill, refine its description to use an invocation-focused third-person format starting with "Use this skill when...". Enforcing gerund naming and progressive disclosure across reference files ensures reliable automatic invocation.

Does maintaining multiple Claude Code skills require specific naming conventions?

Yes, maintaining multiple skills requires enforcing gerund naming conventions and invocation-focused descriptions. This prevents duplication and ensures Claude reliably discovers the right capability when processing CLI and Node.js workflows.

What is the best way to structure an overlong SKILL.md definition?

The best way to structure an overlong SKILL.md is to apply progressive disclosure by shortening the main file into pointers and splitting detailed content into intention-revealing reference files. This maintains reliable automatic invocation.