Verified:GitHub

anthropic-skill-creator

Guide authors in creating modular Claude skills with SKILL.md metadata and instruction bodies.

Updated Aug 27, 2026

Search Tags:#claude #packaging #skill design #progressive disclosure #yaml frontmatter #workflow patterns #instruction authoring

Ready to run

npx skills add https://github.com/mengblom/dex-ghx --skill anthropic-skill-creator-mengblom

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: anthropic-skill-creator
Source: https://github.com/mengblom/dex-ghx/tree/main/.claude/skills/anthropic-skill-creator
Command: npx skills add https://github.com/mengblom/dex-ghx --skill anthropic-skill-creator-mengblom

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill helps you create strong, modular Claude skills by providing a clear framework for writing SKILL.md and organizing optional resources so they load efficiently at runtime.

Core Features & Use Cases

Skill authoring guidance: Explains how to structure a SKILL.md with correct frontmatter (name/description) and an instruction body that stays concise.
Resource design patterns: Recommends when to use scripts, references, and assets, and how to keep SKILL.md lean via progressive disclosure.
Packaging-readiness: Covers practical steps for iterating toward a validated, distributable skill.

Quick Start

Tell the AI: "Create an outline for a new skill called X that includes the required SKILL.md frontmatter, a concise workflow body, and a plan for which resources go in scripts, references, and assets."

Frequently Asked Questions about anthropic-skill-creator

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

FAQPage Schema

How do I structure a SKILL.md file for Claude skills?▼

Structure a SKILL.md file by starting with YAML frontmatter containing the skill name and description, followed by a concise instruction body. Keep operational instructions brief and use progressive disclosure to delegate complex workflows to separate resource files.

What is progressive disclosure in skill design?▼

Progressive disclosure in skill design is the practice of keeping the main instruction body lean by moving detailed logic into modular resources. This approach ensures the SKILL.md loads efficiently at runtime by separating scripts, references, and output assets into distinct files.

When should I use scripts, references, or assets in a Claude skill?▼

Use scripts for executable actions, references for static informational context, and assets for generated output files when packaging a Claude skill. Organizing resources into these distinct categories maintains a lean SKILL.md and ensures efficient runtime loading.

How do I package a Claude skill for distribution?▼

Package a Claude skill for distribution by validating that the SKILL.md has correct frontmatter and a concise workflow body, then organizing all modular resources like scripts and references into a compliant directory structure ready for runtime loading.

Do I need YAML frontmatter to create a Claude skill?▼

Yes, you need YAML frontmatter to create a Claude skill. Skill structure rules require compliant frontmatter containing both the skill name and description fields before the instruction body to ensure proper runtime loading and metadata parsing.

What are the limitations of keeping all instructions in SKILL.md?▼

Keeping all instructions directly in SKILL.md creates runtime loading inefficiencies and bloats the file. Skipping progressive disclosure patterns prevents modular organization of scripts and references, making the skill harder to iterate and validate for distribution.