create-custom-skill

Create and audit VS Code Agent Skill directories and SKILL.md files.

1|Updated Feb 28, 2026
One-click install
npx skills add https://github.com/Binh230199/ai --skill create-custom-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-custom-skill
Source: https://github.com/Binh230199/ai/tree/main/.github/skills/create-custom-skill
Command: npx skills add https://github.com/Binh230199/ai --skill create-custom-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It helps you create, update, and review VS Code Agent Skills without guessing the required folder structure, frontmatter, or validation rules.

Core Features & Use Cases

  • Skill authoring: Build new skills with the correct name, description, and body format.
  • Skill review: Check existing skills for directory-name mismatches, missing sections, and frontmatter issues.
  • Skill maintenance: Keep reusable workflows lean by deciding when to extend an existing skill instead of creating a new one.

Quick Start

Ask the agent to help you create or review a skill by giving the skill name and a short description of the capability you want.

Frequently Asked Questions about create-custom-skill

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

FAQPage Schema
How do I create a VS Code Agent Skill with the correct directory structure and frontmatter?

To create a VS Code Agent Skill, generate a directory in the .github/skills folder with a matching SKILL.md file containing valid YAML frontmatter for name, description, and body content. The skill requires correct directory naming and companion-file validation to ensure proper Copilot discovery.

What is the required structure for a Copilot skill SKILL.md file?

A Copilot skill SKILL.md file requires YAML frontmatter with matching fields, a valid directory name, and companion-file validation. It must include structured sections for the skill name, description, and body format to pass audit checks for reusable workflow packaging.

How do I validate an existing VS Code Agent Skill for frontmatter issues and directory mismatches?

Validate an existing VS Code Agent Skill by running the bundled audit script against the .github/skills folder. This checks for directory-name mismatches, missing sections, and frontmatter issues to ensure custom assistant capabilities meet discovery metadata requirements.

Can I use this skill to review and maintain reusable Copilot workflows in the .github/skills folder?

Yes, you can review and maintain reusable Copilot workflows by auditing skill content in the .github/skills folder. The validation script checks frontmatter fields and directory naming, helping decide when to extend an existing skill instead of creating a new one.

When should I extend an existing VS Code Agent Skill instead of creating a new one?

Extend an existing VS Code Agent Skill instead of creating a new one when the reusable workflow packaging already covers similar custom assistant capabilities. The skill maintenance process helps keep workflows lean by evaluating frontmatter and structure to avoid duplicate skill authoring.

Why does my Copilot skill fail discovery validation due to frontmatter and directory name mismatches?

Copilot skill discovery validation fails when the YAML frontmatter fields do not match the directory name or companion files. The bundled audit script detects these mismatches in the .github/skills folder to ensure correct structure and metadata for agent skill registration.