meta-skill | sys-core | ClaudePluginHub

Skill

Community

meta-skill

Description

Comprehensive guide for creating Claude Code skills following Universal Agentic Runtime standards. MUST Use when building skills, designing skill architecture, implementing progressive disclosure, or validating skill structure.

AI Summary

Provides expert guidance for creating Claude Code skills following Universal Agentic Runtime standards.

Install

Add the repository(one-time)

/plugin marketplace add Git-Fg/thecattoolkit

Install the plugin

/plugin install sys-core@cattoolkit

Tool Access

This skill is limited to using the following tools:

ReadWriteEditGlobGrepBash(mkdir:-p)Bash(ls:*)Bash(cat:*)Bash(find:*)Bash(rm:*)Bash(mv:*)

Supporting Assets

View in Repository

assets/skill-checklist.md

assets/skill-minimal.md

assets/skill-progressive.md

assets/skill-router.md

assets/skill-standard.md

examples/router-pattern-example.md

examples/simple-skill-example.md

references/advanced-patterns.md

references/anti-patterns.md

references/creation-workflow.md

references/description-patterns.md

references/integration-guide.md

references/naming-conventions.md

references/progressive-disclosure.md

references/quick-reference.md

references/router-pattern.md

references/templates-usage.md

references/templates/skill-checklist.md

references/templates/skill-minimal.md

references/templates/skill-progressive.md

Skill Content

Meta-Skill: Skill Architecture Authority

Objective

Meta-Skill provides expert guidance for creating effective Claude Code skills that embody the Universal Agentic Runtime principles. It teaches you how to analyze requirements, choose appropriate templates, implement progressive disclosure, and validate skill structures to ensure optimal performance and maintainability.

Core Principle: Skills are knowledge bases that provide domain expertise through progressive disclosure, not workflow routers. They should be designed for reusability and optimized for context window efficiency.

Context & Standards

Skill creation follows Universal Agentic Runtime standards emphasizing:

Progressive Disclosure: Core content in SKILL.md (< 500 lines), detailed docs in references/
Inline-First: Avoid unnecessary forking (cost: inline=1, fork=3)
Zero-Waste: Maximize inline skills for local engineering tasks
Intent-Driven: Skills are discovered by intent, not called procedurally

Quick Start Algorithm

ANALYZE → Understand purpose, complexity, and use cases
DESIGN → Choose template (Standard/Progressive/Router/Minimal/Checklist)
IMPLEMENT → Create SKILL.md with proper frontmatter
VALIDATE → Run validation scripts
OPTIMIZE → Apply progressive disclosure principles

Detailed Workflow: references/creation-workflow.md

Capability Index & References

1. Creation & Architecture

Creation Workflow: references/creation-workflow.md
Naming Conventions: references/naming-conventions.md
Description Patterns: references/description-patterns.md
Progressive Disclosure: references/progressive-disclosure.md

2. Implementation Guides

Standard Skill Creation: See references/templates-usage.md
Advanced Architectures: See references/integration-guide.md
Template Usage: references/templates-usage.md
Integration Guide: references/integration-guide.md

3. Validation & Quality

Validation Rules: See success_criteria section below
Anti-Patterns: references/anti-patterns.md
Quick Reference: references/quick-reference.md

Templates (Assets)

Available in assets/:

Template	Use When	Complexity
Standard	Single capability, straightforward	Low
Progressive	Complex domain expertise	High
Router	Multiple workflows/paths	Medium
Minimal	Internal tools, utilities	Low
Checklist	Task-oriented with validation	Medium

Validation Commands

Run all validations:

uv run scripts/validate-all.py ./my-skill/

Individual Checks:

Name: ./scripts/validate-skill-name.py my-skill
Description: ./scripts/validate-description.py ./my-skill/
Frontmatter: ./scripts/validate-frontmatter.py ./my-skill/
Token Budget: ./scripts/check-token-budget.py ./my-plugin/

Success Criteria

A well-designed skill has:

Links

GitHub Stats

0 forks

Updated 10 hours ago

Similar Skills

Command Development

10 files

This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.

plugin-dev

55.7k

Agent Development

6 files

This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.

plugin-dev

55.7k

Hook Development

10 files

This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.

plugin-dev

55.7k