meta-structure-organizer

name	meta-structure-organizer
description	Organize features into the right structure: Command, Skill, or Agent. USE WHEN: - "should this be a skill?" - "command or agent?" - "how to structure this feature" - "what type should this be" - "skill vs command" - structure/architecture decisions DO NOT USE WHEN: - Actually implementing (use the generated spec instead) - General coding tasks unrelated to structure design

Structure Organizer

Organize features into the right structure: Command, Skill, or Agent for AI coding assistants.

Quick Reference

Component	Trigger	Reasoning	Execution	Use When
Command	Human /command	None	Fixed procedure	Explicit user trigger needed
Skill	Auto-load on keywords	None	No execution (knowledge)	Domain expertise to share
Agent	Goal assigned	LLM decides	Dynamic, iterative	Multi-step planning required

Workflow Routing

Intent	Workflow
Analyze a feature request	workflows/analyze.md
Generate spec template	workflows/generate-spec.md

Core Resources

Resource	Purpose
Decision Tree	Primary decision logic
Criteria Matrix	Detailed comparison criteria
Boundary Cases	12 common confusions
Combination Patterns	Multi-component architectures
Templates	Spec templates for each type

Combination Patterns

Most real features need multiple component types. Common patterns:

Pattern	Structure	Use When
Command + Agent	Entry → Executor	User triggers complex work
Agent + Skills	Executor + Knowledge	Agent needs domain expertise
Command + Skills	Entry + Knowledge	Procedure needs domain knowledge
Full Stack	Command → Agent → Skills → Tools	Complete feature

See combination-patterns.md for details.

Output

This skill outputs:

1. Diagnosis - Component type (Command / Skill / Agent / Combination)
2. Spec Template - Filled template from references/templates/
3. Rationale - Why this type, why not others

Note: This skill diagnoses what type to use. Implementation is handled by the calling Command.

Platform Support

Platform	Commands	Skills	Agents
Claude Code	.claude/commands/*.md	.claude/skills/*/SKILL.md	Subagent via Task
OpenCode	.opencode/commands/*.md	skills/*/SKILL.md	agents/*.md
Cursor	.cursor/commands/*.md	.cursor/rules/*.mdx	Agent mode