skill-creator

Guide creation and optimization of AI skills with SKILL.md structure.

1|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/91zgaoge/memoh-X --skill skill-creator-91zgaoge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/91zgaoge/memoh-X/tree/main/internal/skills/defaults/skill-creator
Command: npx skills add https://github.com/91zgaoge/memoh-X --skill skill-creator-91zgaoge

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured guide and best practices for creating effective, efficient, and robust AI skills, ensuring they seamlessly extend AI capabilities without unnecessary token consumption.

Core Features & Use Cases

  • Skill Anatomy: Understand the required SKILL.md structure and optional bundled resources (scripts/, references/, assets/).
  • Best Practices: Learn principles like conciseness, progressive disclosure, and setting appropriate degrees of freedom.
  • Development Workflow: Follow a step-by-step process for planning, creating, and iterating on skills.
  • Use Case: You are an AI developer tasked with creating a new skill for processing customer support tickets. This skill will guide you on how to structure your SKILL.md, what information to include, and how to leverage reference files for specific support categories.

Quick Start

Use the skill-creator to learn how to build a new skill for summarizing meeting transcripts.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I structure an AI agent skill to minimize token usage?

To structure an AI agent skill and minimize token usage, define a concise SKILL.md file, apply progressive disclosure for information, and bundle specialized knowledge into reference files instead of the main prompt.

What are the best practices for writing prompt engineering workflows?

Best practices for writing prompt engineering workflows include setting appropriate degrees of freedom, keeping instructions concise, and progressively disclosing information to maximize agent capabilities without excessive token consumption.

How to build a custom AI skill for processing customer support tickets?

Build a custom AI skill for support tickets by structuring a SKILL.md file, defining the processing workflow, and leveraging bundled reference files to handle specific support categories effectively.

What is progressive disclosure in AI skill development?

Progressive disclosure in AI skill development is a design principle that minimizes token usage by providing concise instructions upfront and loading detailed context only when required by the agent.

When should I use bundled resources like scripts and references when building AI skills?

Use bundled resources like scripts, references, and assets when building AI skills to offload detailed specialized knowledge from the main prompt, reducing token consumption and extending agent capabilities.

What is the step by step process for creating and iterating on AI skills?

The step by step process for creating AI skills involves planning the structure, writing the SKILL.md file, integrating necessary workflows, and iterating on the instructions to optimize agent performance.