What is the difference between a command and an agent in a plugin?

Commands are user-invoked slash commands, while agents are specialized sub-entities that Claude Code can invoke to handle specific roles or expert tasks.

How does Claude Code handle component discovery?

Claude Code automatically scans specific root-level directories like commands/, agents/, and skills/ for relevant files (.md or SKILL.md) and registers them automatically.

Why should I use ${CLAUDE_PLUGIN_ROOT}?

It is an environment variable that ensures your plugin remains portable across different operating systems and installation paths by providing a reliable reference to the plugin root.

Can I use custom directory names for my plugin components?

Yes, you can define custom paths in the plugin.json manifest, which will supplement the default auto-discovery directories.

Where should the plugin manifest be located?

The plugin.json manifest must always be located within the .claude-plugin/ directory at the root of your plugin folder.

Claude Code Plugin Architect

Name: Claude Code Plugin Architect
Author: Microck

byMicrock

•

108

Developer Tools

Standardizes the creation and organization of Claude Code plugins using best practices for directory layouts and manifests.

About

This skill provides a comprehensive framework for developing Claude Code plugins, ensuring they follow standardized directory structures and manifest configurations. It guides developers through setting up commands, agents, skills, and hooks while implementing portable path references with ${CLAUDE_PLUGIN_ROOT} for seamless cross-platform installation. By automating component discovery and defining explicit naming conventions, it helps developers build maintainable, scalable, and portable Claude Code extensions that integrate perfectly with the platform's native capabilities.

Key Features

Standardized .claude-plugin/plugin.json manifest configuration
Portable path management using ${CLAUDE_PLUGIN_ROOT}
108 GitHub stars
Consistent naming conventions and architectural best practices
Automatic component discovery for commands, agents, and skills
Event-driven hook configuration and MCP server integration

Use Cases

Converting existing terminal scripts into a structured Claude Code plugin
Troubleshooting manifest configuration errors or component discovery issues
Scaffolding a new Claude Code plugin from scratch with the correct directory layout

About

Key Features

Standardized .claude-plugin/plugin.json manifest configuration
Portable path management using ${CLAUDE_PLUGIN_ROOT}
108 GitHub stars
Consistent naming conventions and architectural best practices
Automatic component discovery for commands, agents, and skills
Event-driven hook configuration and MCP server integration

Use Cases

Converting existing terminal scripts into a structured Claude Code plugin
Troubleshooting manifest configuration errors or component discovery issues
Scaffolding a new Claude Code plugin from scratch with the correct directory layout