skills.rest
vinnie357
vinnie357COMMUNITY
Explore all skills by this author →
GitHub

claude-skills

Creates modular Agent Skills with YAML frontmatter and progressive disclosure.

24|6|Updated Nov 15, 2025
Search Tags:#automation#knowledge management#ai development#agent skills#claude skills#skill design#progressive disclosure
One-click install
npx skills add https://github.com/vinnie357/claude-skills --skill claude-skills-vinnie357
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude-skills
Source: https://github.com/vinnie357/claude-skills/tree/main/claude-code/skills/claude-skills
Command: npx skills add https://github.com/vinnie357/claude-skills --skill claude-skills-vinnie357

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a comprehensive guide for creating modular, self-contained Agent Skills, enabling Claude to dynamically acquire specialized knowledge. It solves the challenge of managing vast amounts of context by implementing progressive disclosure, ensuring efficient token usage while supporting unbounded expertise.

Core Features & Use Cases

  • Skill Structure & Progressive Disclosure: Learn to organize skills with SKILL.md, scripts/, references/, and assets/ to load context only when needed, optimizing token efficiency.
  • Seven-Step Workflow: Follow a structured process from understanding use cases and planning resources to initialization, editing, documentation, validation, and iteration.
  • AI-Centric Writing: Master writing clear, imperative instructions and descriptions tailored for AI consumption, ensuring Claude activates and utilizes skills effectively.
  • Use Case: When developing a new skill for Git operations, use this guide to structure your SKILL.md with core concepts, place detailed conventional commit specifications in references/conventional-commits.md, and include a script for commit analysis in scripts/analyze-commit.sh.

Quick Start

Create a new skill directory named 'my-new-skill' with a 'SKILL.md' file and optional 'scripts', 'references', and 'assets' subdirectories.

Frequently Asked Questions about claude-skills

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

FAQPage Schema
How do I create a modular Agent Skill with progressive disclosure?▼

Create a modular Agent Skill by structuring it with a SKILL.md file, scripts/, references/, and assets/ directories. Progressive disclosure loads context only when needed, optimizing token usage while maintaining access to specialized knowledge across unbounded expertise domains.

What's the seven-step workflow for building an Agent Skill?▼

The seven-step workflow guides you from understanding use cases and planning resources through initialization, editing, documentation, validation, and iteration. This structured process ensures your Skill follows standardized packaging with YAML frontmatter, required name and description fields, and optional license and allowed-tools specifications.

How do I organize references and scripts within a Skill?▼

Organize detailed content in references/ subdirectories (e.g., conventional-commits.md for Git specifications) and automation logic in scripts/ (e.g., analyze-commit.sh for commit analysis). This separation keeps core concepts in SKILL.md while enabling on-demand loading of specialized knowledge.

Can I use progressive disclosure to manage large knowledge bases in Agent Skills?▼

Yes. Progressive disclosure enables efficient token usage by loading context only when activated. This approach solves the challenge of managing vast amounts of context, allowing Claude to dynamically acquire specialized knowledge without overwhelming the token budget.

How do I write AI-centric instructions for an Agent Skill?▼

Write clear, imperative instructions tailored for AI consumption in your SKILL.md. Use domain vocabulary, action verbs, and concrete use-case examples to ensure Claude activates and utilizes your Skill effectively during agent workflows.

What's required in a Skill's YAML frontmatter?▼

Required fields in YAML frontmatter are name and description. Optional fields include license and allowed-tools. This standardized metadata structure ensures your Skill integrates with agent skill packaging and enables proper activation and resource management.

Related Skills

Explore More Software Engineering Skills→
theodrosyimertheodrosyimer

distill

Convert source materials into a single structured project knowledge markdown file.

Community
Intermediate
antriksh-29antriksh-29

qa-only

Generate structured QA reports with health scores, screenshots, and reproducible steps.

Community
Advanced
an8079an8079
4

langchain-arch

Implement LangChain agents, chains, memory, and RAG systems for LLM applications.

Community
Intermediate
VMRam95VMRam95

skill-creator

Create and validate modular AI skills with structured directories and progressive disclosure.

Community
Intermediate