write-a-skill

Create reusable agent skills with SKILL.md and optional resource files.

Updated May 8, 2026

Search Tags:#agent skills #prompt design #skill authoring #progressive disclosure #resource organization #instruction writing

Ready to run

npx skills add https://github.com/mmartinsoliv/dotfiles --skill write-a-skill-mmartinsoliv

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: write-a-skill
Source: https://github.com/mmartinsoliv/dotfiles/tree/main/agents/skills/write-a-skill
Command: npx skills add https://github.com/mmartinsoliv/dotfiles --skill write-a-skill-mmartinsoliv

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you create a new agent skill that is easy to discover, easy to maintain, and clear about when it should be used.

Core Features & Use Cases

Skill Planning: Defines the skill’s purpose, target use cases, and required resources before writing anything else.
Instruction Design: Produces concise SKILL.md guidance with progressive disclosure so the agent gets the right amount of context at the right time.
Resource Organization: Supports optional scripts, references, and assets when the skill needs deterministic actions or deeper documentation.
Use Case: Use it when you want to turn a task domain into a polished, reusable assistant capability with a clean folder structure.

Quick Start

Ask for a new skill draft that includes the skill’s goal, use cases, required files, and any supporting resources.

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 reusable agent skills with structured instructions?▼

To create reusable agent skills, you need a root SKILL.md file with YAML frontmatter and concise operational guidance. This structures capabilities, use cases, and progressive disclosure for assistant behavior.

What is progressive disclosure in skill authoring?▼

Progressive disclosure in skill authoring provides the right amount of context to an agent at the right time. It structures instructions so agents load operational guidance and optional resource files only when needed.

How do I structure optional scripts and references for an assistant capability?▼

You structure optional scripts and references using a clean folder structure for on-demand loading. Organize supporting assets alongside the root SKILL.md so the agent accesses deeper documentation or deterministic actions only when required.

Do I need YAML frontmatter to build a new agent skill?▼

Yes, you need YAML frontmatter in the root SKILL.md file to build a new agent skill. It defines the skill's metadata, ensuring the capability is easy to discover, maintain, and clear about when it should be used.

What is the best way to turn a task domain into a polished assistant capability?▼

The best way to turn a task domain into a capability is defining the skill's purpose, target use cases, and required resources before writing instructions. This planning phase ensures concise guidance and proper resource organization.

When should I not use a single SKILL.md file for skill authoring?▼

You should not rely solely on a single SKILL.md file when your skill needs deterministic actions or deeper documentation. In these cases, add optional scripts, references, and assets organized for on-demand loading.