brainstorm

name	brainstorm
description	Guides collaborative dialogue to turn ideas into designs through one-at-a-time questions, approach exploration with trade-offs, and incremental validation. Does not implement code. Use before creative work or significant changes — when the user asks to "brainstorm", "think through", "explore options for", or similar.
user-invocable	true
disable-model-invocation	false

Brainstorm

Exploratory dialogue — same shape as /add-feature Phases 1–5, minus implementation. Use this when the user wants to explore before committing to a direction.

Phase 1: Understand

Read AGENTS.md and the files in example/src/ relevant to the idea. Find similar existing patterns. Map out which files or behaviors would be affected.

Phase 2: Clarify (one question at a time)

Ask focused questions one at a time. Prefer multi-choice. Topics:

• What problem is this actually solving?
• Who is the user?
• What are the scope boundaries?
• Any constraints (performance, accessibility, bundle size, CRA limits)?

One question per message. Do not batch.

Phase 3: Explore Approaches

Propose 2–3 approaches with trade-offs. Lead with a recommendation.

**Option A: [name]** (recommended)
- how it works: …
- pros: …
- cons: …

**Option B: [name]**
- how it works: …
- pros: …
- cons: …

Which direction appeals to you?

Phase 4: Design (incremental validation)

Once a direction is chosen, present the shape in 200–300 word sections — component placement, state, data flow, error cases, tests. Ask after each section before continuing.

Phase 5: Exit

Use AskUserQuestion to pick what's next — this skill does not implement:

{
  "questions": [{
    "question": "Brainstorm complete. What's next?",
    "header": "Next step",
    "options": [
      {"label": "Write plan file", "description": "Save the design as a plan document for later execution"},
      {"label": "Enter plan mode", "description": "Use native plan mode with an approval gate before implementation"},
      {"label": "Start implementing", "description": "Switch to /add-feature and begin"}
    ],
    "multiSelect": false
  }]
}

Principles

• Exploration only. Do not write implementation code in this skill.
• One question at a time.
• Always present 2–3 approaches with trade-offs before narrowing down.
• YAGNI — cut scope ruthlessly before committing.
• Lead with a recommendation; let the user redirect.