write-a-skill

Create AI agent skills with SKILL.md, YAML frontmatter, and optional supporting files.

1|Updated May 6, 2026
One-click install
npx skills add https://github.com/ThorStarlord/interface-skills --skill write-a-skill-thorstarlord
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-a-skill
Source: https://github.com/ThorStarlord/interface-skills/tree/main/skills/write-a-skill
Command: npx skills add https://github.com/ThorStarlord/interface-skills --skill write-a-skill-thorstarlord

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps teams consistently author new AI agent skills with clear purpose, reliable structure, and reusable bundled resources.

Core Features & Use Cases

  • Requirements gathering: Captures domain, use cases, and whether scripts or reference material are needed to meet the user’s intent.
  • Skill drafting with correct scaffolding: Produces a valid SKILL.md plus optional supporting files (references, examples, and scripts) without bloating the core entrypoint.
  • Review-driven refinement: Uses an explicit checklist to ensure the description includes actionable triggers and stays within size/quality constraints.

Quick Start

Ask the agent to draft a new skill named "my-skill" with a trigger-based SKILL.md description, a concise workflow, and optional REFERENCE.md if the content is long.

Frequently Asked Questions about write-a-skill

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

FAQPage Schema
How do I structure an AI agent skill with YAML frontmatter and progressive disclosure?

Structuring an AI agent skill requires a root SKILL.md with YAML frontmatter for name and description, using progressive disclosure to move detailed content into optional REFERENCE.md and EXAMPLES.md files to keep the core entrypoint concise.

What is the best way to author agent workflows with bundled resources and scripts?

Authoring agent workflows with bundled resources involves drafting a SKILL.md with operational instructions, then adding deterministic scripts and reference documentation only when the specific domain requires them to meet user intent.

How do I write a trigger-based description for a new skill package?

Writing a trigger-based skill description involves using a review checklist to ensure the YAML frontmatter description includes actionable triggers and stays within size constraints so the agent knows exactly when to activate the workflow.

When do I need separate REFERENCE.md and EXAMPLES.md files for skill authoring?

You need separate REFERENCE.md and EXAMPLES.md files for skill authoring when the domain content is too long for the root SKILL.md, using progressive disclosure to prevent bloating the core operational instructions.

Can I include deterministic scripts in my skill package without bloating the SKILL.md file?

You can include deterministic scripts in your skill package by bundling them as optional supporting files alongside the root SKILL.md, keeping the core entrypoint clean while supporting the defined workflow.