write-a-skill

Scaffold SKILL.md files with YAML frontmatter and concise bodies.

1|Updated Jun 9, 2026
One-click install
npx skills add https://github.com/prestarius/jet-skills-stack --skill write-a-skill-prestarius
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-a-skill
Source: https://github.com/prestarius/jet-skills-stack/tree/main/skills/write-a-skill
Command: npx skills add https://github.com/prestarius/jet-skills-stack --skill write-a-skill-prestarius

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Author a skill that triggers reliably and stays small by ensuring correct structure, frontmatter, and progressive disclosure.

Core Features & Use Cases

  • Scaffold the files: Create skills/<kebab-name>/SKILL.md with YAML frontmatter name and description and a concise body.
  • Apply progressive disclosure: Move long detail into bundled references/, assets/, or scripts/ and refer to them from the body.
  • Wire-in guidance: Update repository README and follow house conventions for a company-agnostic skill.

Quick Start

Create skills/<kebab-name>/SKILL.md with frontmatter including name and description, and implement a concise, deterministic workflow in the body.

Frequently Asked Questions about write-a-skill

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

FAQPage Schema
How do I create a Claude Code skill with correct SKILL.md frontmatter?

To create a Claude Code skill, you scaffold a skills/<kebab-name>/SKILL.md file with YAML frontmatter containing name and description, then implement a concise, deterministic workflow in the body.

What is progressive disclosure in Claude Code skills?

Progressive disclosure in Claude Code skills moves long detail into bundled references/, assets/, or scripts/ directories, keeping the SKILL.md body concise while referring to those external files for extended context.

How do I validate frontmatter for a new Claude Code skill?

Validating frontmatter for a Claude Code skill involves checking that the YAML header includes well-defined name and description fields, ensuring the structure triggers reliably and adheres to house conventions.

Can I package repeatable workflows into a Claude Code skill?

Yes, you can package repeatable workflows into a Claude Code skill by authoring a concise, deterministic body in SKILL.md, moving lengthy details into bundled references, and updating the repository README.

What's the best way to keep a Claude Code skill compact and reliable?

The best way to keep a Claude Code skill compact is enforcing a concise body paired with minimal frontmatter, applying progressive disclosure to externalize details, and following consistent file scaffolding conventions.

Why does my Claude Code skill not trigger reliably?

A Claude Code skill may not trigger reliably if the SKILL.md frontmatter is malformed or missing required fields, or if the body is overly verbose instead of concise and deterministic.