plugin-dev

Guided end-to-end plugin creation workflow with component design, implementation, and validation

**Effective agents are focused tools, not Swiss Army knives.** The best agents do one thing well with clear triggers and minimal tooling.

You are an expert plugin validator specializing in comprehensive validation of Claude Code plugin structure, configuration, and components.

**Effective skills trigger reliably and deliver focused value.** The three quality signals that matter most: triggering, content density, and progressive disclosure.

Guides development of Claude Code plugin agents, including agent file structure, YAML frontmatter fields, system prompt design, triggering conditions, and model/color/tool configuration. Activates when the user asks to create an agent, add a subagent, write agent frontmatter, configure agent triggering examples, choose agent tools or colors, or needs guidance on agent development best practices.

Guides development of Claude Code slash commands, including file format, YAML frontmatter fields, dynamic arguments ($ARGUMENTS, $1/$2), file references, bash execution, and command organization. Activates when the user asks to create a slash command, add a command, define command arguments, use command frontmatter, organize commands, create commands with file references, or needs guidance on command development best practices.

This skill should be used when the user asks to "create a plugin", "build a new plugin", "scaffold a plugin", "make a Claude Code plugin", "design a plugin", or needs guidance on the end-to-end plugin creation workflow including discovery, component planning, design, implementation, validation, and testing.

Guides development of Claude Code plugin hooks, including prompt-based and command hooks, hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification), matchers, output formats, and security practices. Activates when the user asks to create a hook, add event-driven automation, validate tool use, implement prompt-based hooks, block dangerous commands, or configure hooks.json.

Guides integration of Model Context Protocol (MCP) servers into Claude Code plugins, including server types (stdio, SSE, HTTP, WebSocket), .mcp.json configuration, authentication patterns (OAuth, tokens), environment variable expansion, and MCP tool naming. Activates when the user asks to add an MCP server, configure .mcp.json, connect an external service, or set up MCP in a plugin.