create-skill

Create reusable Claude Code skills with YAML frontmatter and markdown instructions.

514|191|Updated Mar 28, 2017
One-click install
npx skills add https://github.com/tidyverse/dbplyr --skill create-skill-tidyverse
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-skill
Source: https://github.com/tidyverse/dbplyr/tree/main/.claude/skills/create-skill
Command: npx skills add https://github.com/tidyverse/dbplyr --skill create-skill-tidyverse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill empowers you to package complex, multi-step workflows or specialized domain expertise into reusable AI capabilities. It streamlines the process of defining new skills, ensuring consistency and reducing the manual effort involved in teaching an AI new tasks.

Core Features & Use Cases

  • Structured Skill Definition: Guides you through creating a SKILL.md file with proper YAML frontmatter and clear Markdown instructions, ensuring your skill is discoverable and actionable.
  • Workflow Encoding: Helps you formalize repeating workflows, specific domain knowledge (like SQL translation patterns), or consistent processes into an autonomous capability.
  • Use Case: If you frequently find yourself explaining a specific coding pattern, a multi-step deployment process, or a data analysis routine to an AI, use this skill to formalize that knowledge into a reusable, auto-invocable skill.

Quick Start

Create a new skill named 'code-review' that provides a checklist for reviewing Python code for common security issues.

Frequently Asked Questions about create-skill

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

FAQPage Schema
How do I package workflows and expertise into reusable AI skills?

Package workflows into reusable AI skills by creating a SKILL.md file with YAML frontmatter and markdown instructions that define repeatable tasks, domain knowledge, or multi-step processes. This structured format enables the skill to be discovered, invoked, and applied consistently across projects without manual re-explanation.

What's the proper structure for defining a custom AI skill?

A custom AI skill requires a SKILL.md file with YAML frontmatter specifying metadata and clear markdown instructions describing the task logic. This enforces naming conventions and consistency standards, ensuring your skill is discoverable and can be automatically invoked by Claude when relevant.

Can I automate repeated coding patterns or deployment processes as skills?

Yes, you can formalize recurring coding patterns, deployment sequences, or domain-specific routines into autonomous skills. This eliminates repetitive explanation and enables consistent execution of complex workflows, such as security review checklists or infrastructure provisioning steps, through a single reusable capability.

How do I ensure my skill is discoverable and consistent?

Enforce discoverability and consistency by following the SKILL.md structure with required YAML frontmatter, naming conventions, and optional supporting files. This standardized format integrates your skill into Claude's capability registry, making it automatically available when matching user intent.

What types of knowledge can I encode into a skill?

Encode specialized domain knowledge, multi-step workflows, coding patterns, analysis routines, or consistent processes into skills. Whether it's SQL translation patterns, security review procedures, or data workflows, skills transform knowledge you repeatedly teach into autonomous, invocable capabilities.