mcpField
C
create-skill

create-skill

github.com/jasonharmongitAgent workflows38 installs
Repository

About create-skill

A guide skill for authoring agent skills in the SKILL.md format. It explains required frontmatter, directory layout, and concise instruction-writing principles for skill creators.

Key features

  1. 01Define YAML frontmatter fields
  2. 02Organize skill files and folders
  3. 03Apply concise authoring principles

Use cases

  1. 01Draft a new SKILL.md file
  2. 02Structure references and example files
  3. 03Trim instructions for agent context

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/jasonharmongit/jh-skills --skill create-skill

After installing, type/create-skillin Claude Code to confirm it loaded.

Details

From the project README

Create Skill

Skill Anatomy

Frontmatter

Every SKILL.md file starts with YAML frontmatter:

---
name: your-skill-name
description: Brief high-level description of what this skill does.
disable-model-invocation: true
---
  • name: Kebab-case identifier for the skill
  • description: A single sentence giving a very high-level summary of what the skill does. Do not include specifics about how the skill works, when to invoke it, or trigger terms.
  • disable-model-invocation: Include disable-model-invocation: true by default so the skill only loads when explicitly named.

Skill File Structure

Directory Layout

Skills are stored as directories containing a SKILL.md file:

skill-name/
├── SKILL.md              # Required - main instructions
├── reference.md          # Optional - detailed documentation
├── examples.md           # Optional - usage examples
└── scripts/              # Optional - utility scripts
    ├── validate.py
    └── helper.sh

Core Authoring Principles

1. Concise is Key

The context window is shared with conversation history, other skills, and requests. Every token competes for space.

Default assumption: The agent is already very smart. Only add context it doesn't already have.

Challenge each piece of information:

  • "Does the agent really need this explanation?"
  • "Can I assume the agent knows this?"
  • "Does this paragraph justify its token cost?"

Comments

Markdown supported