skill-creator

Standardize SKILL.md metadata and progressive disclosure for Claude skill directories.

1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/LaunchPadLab/ai-bank --skill skill-creator-launchpadlab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/LaunchPadLab/ai-bank/tree/main/claude/skills/skill-creator
Command: npx skills add https://github.com/LaunchPadLab/ai-bank --skill skill-creator-launchpadlab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires PyYAML, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Skill-creator solves the problem of building effective, reusable Claude skills by giving clear guidance on structure, metadata, and progressive disclosure so the right context is loaded at the right time.

Core Features & Use Cases

  • Skill design guidance: Explains how to structure SKILL.md frontmatter and instructional bodies for reliable triggering and context efficiency.
  • Progressive disclosure patterns: Teaches a staged loading approach (metadata first, then body on trigger, then optional references/resources) to keep context windows lean.
  • End-to-end workflow: Covers the practical lifecycle from understanding examples through planning resources, initializing, editing, and packaging a skill.
  • Use case: When you need to create or update a domain-specific skill (e.g., for a workflow, tool integration, or reusable templates/scripts), you use this guidance to produce a consistent package.

Quick Start

Use the skill-creator guidance to draft your new SKILL.md frontmatter (name and description) and then write the essential operational instructions in the body while linking to any additional resources you add.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create a Claude skill with proper YAML frontmatter and metadata?

To create a Claude skill, you standardize SKILL.md metadata by defining required YAML frontmatter fields like name and description, then write operational instructions in the body to ensure the directory validates and loads deterministically.

What is progressive disclosure in skill design and why does it matter?

Progressive disclosure in skill design is a staged context loading approach where metadata loads first, the instructional body loads on trigger, and optional references load last, keeping the context window lean and ensuring the right context is loaded at the right time.

Do I need PyYAML to package and validate deterministic Claude workflows?

Yes, PyYAML is a required dependency to package and validate deterministic Claude workflows, ensuring repository-safe organization and proper parsing of YAML frontmatter fields within the SKILL.md file structure.

How do I structure a skill directory to bundle scripts and references for recurring tasks?

You structure a skill directory by organizing bundled scripts, references, and assets alongside a SKILL.md file, applying repository-safe organization principles so the specialized knowledge and tool integrations validate and load deterministically for recurring tasks.

What's the best way to update an existing Claude skill without breaking context management?

The best way to update an existing Claude skill is to edit the SKILL.md instructional body and linked resources while maintaining progressive disclosure patterns, ensuring context management remains efficient and the updated directory loads deterministically.

When should I not use progressive disclosure for skill design?

You should avoid progressive disclosure for skill design when your operational instructions are minimal enough to fit entirely within the SKILL.md body without overwhelming the context window, making staged loading of optional references unnecessary.