What is the difference between a Skill and a Hook?

Hooks are for event-driven automation and guardrails, while Agent Skills are reusable playbooks for specific, deterministic tasks.

Does this skill help with testing?

Yes, it includes an evaluation strategy phase that helps you define 20-50 test cases to validate your extension's performance.

When should I use an Agent Team?

Agent Teams are recommended for high-complexity tasks (score 9-10) involving multiple specialized domains and complex coordination.

How does the complexity scoring work?

The skill uses a 0-10 scale: scores of 1-3 suggest Simple Skills, 4-6 for Workflows, 7-8 for Sub-Agents, and 9-10 for complex Agent Teams.

Claude Extension Architect

Name: Claude Extension Architect
Author: yu-iskw

byyu-iskw

0•

Herramientas para Desarrolladores

Guides the selection and implementation of Claude extensions by matching task complexity to the ideal architectural pattern.

The Claude Extension Architect skill serves as a strategic orchestration layer for developers looking to extend Claude's capabilities. It utilizes a rigorous complexity-based scoring framework (0-10) to help users navigate the choice between Hooks, Agent Skills, Sub-Agents, Agent Teams, and Plugins. By providing a structured workflow that includes complexity assessment, validation, and lifecycle checks, this skill ensures that developers build the simplest possible solution for their requirements while maintaining alignment with Anthropic's recommended best practices and evaluation strategies.

Características Principales

01Complexity-based decision framework for architectural selection

020 GitHub stars

03Standardized lifecycle management from design to verification

04Integrated evaluation strategies with 20-50 test case planning

05Comparative analysis of Hooks, Skills, Sub-Agents, and Agent Teams

06Automated routing to specific component implementation workflows

Casos de Uso

01Standardizing the creation and packaging of internal Claude plugins

02Determining the most efficient way to automate a specific developer workflow in Claude Code

03Scaling simple agent skills into complex multi-agent teams for enterprise tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yu-iskw/claude-plugin-builder implement-claude-extensions

For use in Claude.ai and ChatGPT

Características Principales

01Complexity-based decision framework for architectural selection

020 GitHub stars

03Standardized lifecycle management from design to verification

04Integrated evaluation strategies with 20-50 test case planning

05Comparative analysis of Hooks, Skills, Sub-Agents, and Agent Teams

06Automated routing to specific component implementation workflows

Casos de Uso

01Standardizing the creation and packaging of internal Claude plugins

02Determining the most efficient way to automate a specific developer workflow in Claude Code

03Scaling simple agent skills into complex multi-agent teams for enterprise tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yu-iskw/claude-plugin-builder implement-claude-extensions

For use in Claude.ai and ChatGPT