skills.rest
jubalm
jubalmCOMMUNITY
Explore all skills by this author →
GitHub

Authoring Agent Skills

Create reusable Claude Skills with YAML metadata and Markdown instructions.

1|Updated Oct 1, 2023
Search Tags:#skill development#custom skills#workflow automation#knowledge management#ai skills#agent authoring#claude customization
One-click install
npx skills add https://github.com/jubalm/dotfiles --skill authoring-agent-skills
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Authoring Agent Skills
Source: https://github.com/jubalm/dotfiles/tree/main/.claude/skills/authoring-agent-skills
Command: npx skills add https://github.com/jubalm/dotfiles --skill authoring-agent-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solves? This Skill empowers you to build specialized, reusable AI capabilities for Claude, eliminating repetitive instructions and streamlining complex workflows. It transforms Claude from a generalist into a domain-specific expert tailored to your needs.

Core Features & Use Cases

  • Structured Skill Creation: Learn to define Skills with YAML metadata and Markdown instructions for clear, discoverable functionality.
  • Progressive Disclosure: Organize complex knowledge into modular files, ensuring Claude loads only relevant information on demand, optimizing context usage.
  • Use Case: Create a "PDF Text Extraction" Skill that Claude can automatically invoke whenever you mention "PDF" or "text extraction," providing specific Python code snippets for the task.

Quick Start

  1. Create a directory for your skill, e.g., ~/.claude/skills/my-pdf-skill/

  2. Inside, create SKILL.md with frontmatter and instructions:

    name: PDF Text Extraction description: Extract text from PDFs. Use when working with PDF files or text extraction.

    PDF Text Extraction

    Quick start

    import pdfplumber with pdfplumber.open("file.pdf") as pdf: text = pdf.pages[0].extract_text()
  3. Test by mentioning "PDF" or "text extraction" to Claude.