create-ai-tool

Classify user requirements as Skill or Agent and output metadata YAML.

1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/khw1031/ai-library --skill create-ai-tool-khw1031
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-ai-tool
Source: https://github.com/khw1031/ai-library/tree/main/.claude/skills/create-ai-tool
Command: npx skills add https://github.com/khw1031/ai-library --skill create-ai-tool-khw1031

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps analysts determine the correct AI tool type (Skill or Agent) for a given requirement and guides the user to the appropriate creation path.

Core Features & Use Cases

  • Decision guidance that classifies a user request as Skill or Agent.
  • Clear path routing to create-skill or create-agent, with links to the relevant SKILL.md entries.
  • Reference to the decision-tree.md for detailed criteria and edge-case handling.

Quick Start

Analyze a user request to determine whether to create a Skill or an Agent, then direct to the appropriate creation path.

Frequently Asked Questions about create-ai-tool

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

FAQPage Schema
How do I decide whether to build an AI Skill or an Agent for my workflow?

To decide between an AI Skill or an Agent, you classify user requirements using a decision tree that validates prompt metadata and outputs a structured YAML guiding the creation path.

What is the difference between a Skill and an Agent in prompt engineering?

The difference between a Skill and an Agent in prompt engineering depends on tool type classification, where Skills handle specific guided tasks and Agents manage broader autonomous workflows based on requirement analysis.

How do I validate required metadata when designing an AI tool prompt?

To validate required metadata when designing an AI tool prompt, check the SKILL.md file for the presence of a name and description, evaluate optional components, and output a structured metadata YAML.

Can I use a decision tree to classify AI tooling requests?

Yes, you can use a decision tree to classify AI tooling requests by analyzing user specifications for what to build and how it should be invoked, routing them to the correct creation path.

What is the best way to structure metadata for an AI creation tool?

The best way to structure metadata for an AI creation tool is to generate a structured metadata YAML after validating the presence of required name and description fields alongside optional components.

When should I not use a Skill for my AI workflow design?

You should not use a Skill for your AI workflow design when the user requirement classification process determines the request involves autonomous decision-making, routing it to the Agent creation path instead.