Plugin Structure

Design and validate Claude Code plugin directory structures and manifests.

432|40|Updated Jan 11, 2026
One-click install
npx skills add https://github.com/tzachbon/smart-ralph --skill plugin-structure-tzachbon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Plugin Structure
Source: https://github.com/tzachbon/smart-ralph/tree/main/.agents/skills/Plugin%20Structure
Command: npx skills add https://github.com/tzachbon/smart-ralph --skill plugin-structure-tzachbon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps users understand and implement a clear plugin structure for Claude Code, including manifest placement, component discovery, and consistent organization across commands, agents, skills, hooks, and MCP configurations.

Core Features & Use Cases

  • Standardized layout: Provides a consistent blueprint for plugin directories (commands, agents, skills, hooks, and scripts).
  • Manifest-driven integration: Explains how to configure .claude-plugin/plugin.json and optional paths for discovery.
  • Guided organization: Offers best practices for naming, reasoning about component locations, and multi-layer structure to scale plugins.
  • Use Case: You are building a new plugin; this skill guides you to arrange directories, place SKILL.md files, and enable automatic discovery.

Quick Start

To begin, create a plugin skeleton with a manifest at .claude-plugin/plugin.json and add a skills/PLUGIN_STRUCTURE/SKILL.md describing the skill.

Frequently Asked Questions about Plugin Structure

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

FAQPage Schema
How do I structure a Claude Code plugin for automatic component discovery?

To structure a Claude Code plugin for automatic discovery, place your manifest at .claude-plugin/plugin.json and organize components into dedicated directories like commands/, agents/, and skills/ with kebab-case naming.

What is the correct directory layout for Claude Code commands, agents, and hooks?

The correct Claude Code directory layout separates components into distinct folders for commands, agents, skills, hooks, assets, scripts, and MCP configurations, using a manifest-driven approach to define optional discovery paths.

Does a Claude Code plugin require a manifest file and SKILL.md frontmatter?

Yes, a Claude Code plugin requires a manifest at .claude-plugin/plugin.json. Skills require a SKILL.md file with frontmatter containing both name and description to ensure proper validation and discovery.

Can I use custom naming conventions for Claude Code plugin directories?

Claude Code plugin structures prescribe kebab-case naming conventions for directories and files. This standardization ensures portable CLAUDE_PLUGIN_ROOT references and reliable auto-discovery of commands, agents, and hooks.

What's the best way to refactor an existing Claude Code plugin layout?

The best way to refactor a Claude Code plugin layout is to validate directory structures against standardized blueprints, ensure manifest placement at .claude-plugin/plugin.json, and migrate components to kebab-case naming.

Why does my Claude Code plugin component discovery fail after directory changes?

Claude Code plugin discovery fails when components are misplaced outside designated directories, SKILL.md frontmatter lacks required name and description fields, or naming conventions do not follow kebab-case rules.