What is Progressive Disclosure in skill design?

It is a three-level loading system (Metadata > SKILL.md > Resources) that ensures Claude only loads detailed information when it is relevant to the current task.

When should I use scripts instead of text instructions?

Use scripts for tasks that require deterministic reliability, are repeatedly rewritten by the AI, or are computationally complex for an LLM to perform manually.

Why does the Skill Creator emphasize conciseness?

The context window is a shared resource; concise skills ensure Claude has enough room for conversation history and user requests while minimizing token costs.

What is a Claude Code Skill?

A skill is a modular package that extends Claude's capabilities with specialized knowledge, tools, and workflows, transforming it into a domain-specific agent.

Skill Creator

Name: Skill Creator
Author: willasas

bywillasas

0•

学習とドキュメンテーション

Provides a standardized framework and best practices for developing modular, token-efficient capabilities for Claude Code.

Skill Creator is a meta-skill designed to guide developers through the end-to-end process of building, optimizing, and packaging specialized Claude Code skills. It emphasizes the 'Progressive Disclosure' design principle to minimize context window bloat while providing Claude with the precise procedural knowledge, scripts, and domain expertise required for complex tasks. By following the Skill Creator framework, you can transform Claude from a general-purpose assistant into a specialized agent capable of handling company-specific logic, complex API integrations, and deterministic workflows.

主な機能

010 GitHub stars

02Standardized skill architecture including SKILL.md, scripts, and reference assets

03Framework for defining agent 'Degrees of Freedom' based on task fragility

04Token-optimization strategies using progressive disclosure patterns

05Best practices for managing external resources like scripts and documentation

06Guided 6-step skill development lifecycle from planning to iteration

ユースケース

01Optimizing existing skills to reduce token consumption and improve reliability

02Building reusable toolkits for repetitive tasks like API management or data processing

03Creating custom domain-specific agents for proprietary company workflows

主な機能

010 GitHub stars

02Standardized skill architecture including SKILL.md, scripts, and reference assets

03Framework for defining agent 'Degrees of Freedom' based on task fragility

04Token-optimization strategies using progressive disclosure patterns

05Best practices for managing external resources like scripts and documentation

06Guided 6-step skill development lifecycle from planning to iteration

ユースケース

01Optimizing existing skills to reduce token consumption and improve reliability

02Building reusable toolkits for repetitive tasks like API management or data processing

03Creating custom domain-specific agents for proprietary company workflows