skill-developer

Document and manage Claude Code skills with SKILL.md and skill-rules.json.

Updated Nov 29, 2025
One-click install
npx skills add https://github.com/thimslugga/agent-skills --skill skill-developer-thimslugga
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-developer
Source: https://github.com/thimslugga/agent-skills/tree/main/skills/agent/skill-developer
Command: npx skills add https://github.com/thimslugga/agent-skills --skill skill-developer-thimslugga

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This guide provides a comprehensive framework for creating and managing Claude Code skills that align with Anthropic best practices, enabling reliable auto-activation, consistent structure, and safe, transparent governance across skills.

Core Features & Use Cases

  • Clear skill structure and frontmatter guidance for SKILL.md, including name and description.
  • Coverage of trigger types (keywords, intent patterns, file paths, content patterns) and enforcement levels (block, suggest, warn).
  • Guidance on hook mechanisms, session tracking, and progressive disclosure with reference materials.

Quick Start

Create a new skill by adding .claude/skills/{skill-name}/SKILL.md with frontmatter, then register it in .claude/skills/skill-rules.json and test its triggers.

Frequently Asked Questions about skill-developer

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

FAQPage Schema
How do I create a Claude Code skill that auto-activates reliably?

To create a Claude Code skill, add a SKILL.md file with frontmatter to .claude/skills/{skill-name}/, register it in skill-rules.json, and define trigger patterns like keywords or file paths to ensure reliable auto-activation.

What are the trigger types and enforcement levels for Claude Code skills?

Claude Code skills support trigger types including keywords, intent patterns, file paths, and content patterns. Enforcement levels are block, suggest, and warn, allowing transparent governance over skill behavior.

How does progressive disclosure work when managing Claude Code skills?

Progressive disclosure in Claude Code skills uses reference materials to keep the SKILL.md concise while providing deeper context. This aligns with the 500-line rule to maintain efficient, discoverable, and manageable skill structures.

What is the 500-line rule for SKILL.md files?

The 500-line rule limits the SKILL.md file size to ensure efficient parsing and reliable auto-activation. Detailed information is moved into reference materials using progressive disclosure rather than bloating the main skill file.

Can I enforce guardrails using Claude Code skills?

Yes, you can enforce guardrails using block, suggest, and warn enforcement levels. These are configured in skill-rules.json alongside hook mechanisms and session tracking to ensure safe, transparent governance across all skills.

Do I need dependencies to start developing Claude Code skills?

No dependencies are required to start developing Claude Code skills. You only need to create the .claude/skills directory structure, write SKILL.md with proper frontmatter, and register triggers in skill-rules.json.