What problem does it solve? Building agents for Claude Code plugins requires precise frontmatter fields, triggering descriptions, and system prompt structure, and mistakes in any of these cause agents to trigger unreliably or behave unpredictably. ## Core Features & Use Cases - Agent File Authoring: Provides the complete agent markdown format with required frontmatter fields (name, description, model, color, tools) and naming/length validation rules. - Triggering Design: Guides writing prose trigger descriptions and "When to invoke" scenario sections so agents dispatch reliably for both explicit and proactive requests. - System Prompt Patterns: Supplies templates for analysis, generation, validation, and orchestration agents, plus an AI-assisted generation prompt that outputs JSON ready to convert into agent files. - Use Case: You want an agent that reviews code after each feature is written. Use this Skill to generate the agent file, define its trigger scenarios, and validate it with the included script before adding it to your plugin's agents directory. ## Quick Start Create a code review agent for my Claude Code plugin that triggers after new code is written and validate the resulting agent file.
