When should I use a Subagent instead of a Skill?

Use a Subagent for tasks requiring context isolation or parallel execution where only a summary is needed; use a Skill for sharing reusable instructions or workflows within your main conversation.

What is a Claude Code Plugin?

A plugin is a packaging layer that bundles skills, hooks, subagents, and MCP servers into a single installable unit for easy distribution and reuse across different repositories.

How do I get structured JSON output from Claude Code?

By using the programmatic execution mode with the specific output format flags detailed in this skill, you can force Claude Code to return machine-readable JSON.

Can I run Claude Code in a CI/CD pipeline?

Yes, this skill covers programmatic usage using the Agent SDK CLI for headless mode, allowing for automated testing, linting, and deployment tasks in your CI/CD environment.

What is the difference between CLAUDE.md and a Skill?

CLAUDE.md provides persistent context loaded automatically every session for project-specific rules, while Skills are reusable knowledge or workflows triggered on demand or via slash commands.

Claude Code Extensibility Guide

Name: Claude Code Extensibility Guide
Author: Aznatkoiny

byAznatkoiny

0•

開発者ツール

Provides architectural guidance and implementation patterns for extending Claude Code with custom skills, plugins, and automated workflows.

This skill serves as a comprehensive reference for developers looking to customize and scale their Claude Code environment. It offers deep dives into the functional differences between CLAUDE.md, skills, and MCP servers, while providing actionable advice on implementing subagents, agent teams, and custom output styles. Whether you need to run Claude Code programmatically in a CI/CD pipeline, manage context costs, or build complex multi-agent systems, this skill provides the decision frameworks and configuration patterns necessary to optimize your AI-driven development lifecycle.

主な機能

01Documentation for programmatic Claude Code execution and CI/CD integration

02Best practices for packaging extensions into distributable plugins

030 GitHub stars

04Guidelines for creating custom output styles and structured JSON responses

05Architecture patterns for orchestrating Subagents and Agent Teams

06Comparison frameworks for choosing between Skills, MCP, and Subagents

ユースケース

01Deciding on the most efficient way to provide persistent project context to Claude

02Integrating Claude Code into headless environments or automated testing pipelines

03Automating code reviews or deployment tasks through custom slash commands

主な機能

01Documentation for programmatic Claude Code execution and CI/CD integration

02Best practices for packaging extensions into distributable plugins

030 GitHub stars

04Guidelines for creating custom output styles and structured JSON responses

05Architecture patterns for orchestrating Subagents and Agent Teams

06Comparison frameworks for choosing between Skills, MCP, and Subagents

ユースケース

01Deciding on the most efficient way to provide persistent project context to Claude

02Integrating Claude Code into headless environments or automated testing pipelines

03Automating code reviews or deployment tasks through custom slash commands