Update to Compound Writing

{
  "name": "cw-brainstorm",
  "description": "Help users surface raw material for a piece when they don't have a specific idea yet. Invoke with /cw-brainstorm or when user says they want to explore ideas.",
  "included_files": [],
  "skill_md_contents": "---\nname: cw-brainstorm\ndescription: Help users surface raw material for a piece when they don't have a specific idea yet. Invoke with /cw-brainstorm or when user says they want to explore ideas.\n---\n\n# Brainstorm\n\n## Purpose\n\nHelp the user surface raw material for a piece when they arrive without a specific idea. The goal is to uncover something worth writing about through exploratory conversation.\n\n## Floor Criteria\n\nAll ideas must pass the **New, True, Useful** test:\n\n### New\nThe idea either:\n- Makes the new familiar (introduces something unfamiliar and makes it accessible)\n- Makes the familiar new (takes something known and reveals a fresh angle or connection)\n\n**Test:** \"Does this make the new familiar, or the familiar new?\"\n\n### True\nDepending on subject matter, the idea should be:\n- Factually accurate (verifiable claims, correct data)\n- Honest to the writer's experience (authentic, not performed)\n- Resonant to human experience (rings true, recognizable)\n- Logically sound (the argument holds up under scrutiny)\n\n### Useful\nIn order of importance (customizable per writer or piece):\n1. **Generative** — sparks further thinking, opens new questions\n2. **Reframing** — changes how reader sees a problem or situation\n3. **Clarifying** — helps reader understand something they couldn't before\n4. **Actionable** — reader can do something with it\n\n## What to Hunt For\n\nProbe for material that exhibits these qualities:\n\n| Quality | What You're Looking For |\n|---------|-------------------------|\n| **Surprise** | Something that defied expectations |\n| **Tension** | Contradictions, paradoxes, things that shouldn't both be true |\n| **Transformation** | Change over time—in thinking, behavior, understanding |\n| **Friction** | Frustrations, obstacles, things that drive the user crazy |\n| **Pattern recognition** | Connections across domains the user has noticed |\n| **Earned insight** | Something learned through experience, not just reading |\n\n## How to Conduct the Session\n\n1. **Start open-ended.** Don't lead with a framework—let the user talk.\n\n2. **Ask follow-up questions** that dig into the qualities above:\n   - \"What surprised you about that?\"\n   - \"What changed for you?\"\n   - \"What do most people get wrong about this?\"\n   - \"Why does this matter to you personally?\"\n   - \"What's the tension here?\"\n\n3. **Test ideas against the floor.** When something promising surfaces, mentally check:\n   - Is it new (or a new angle)?\n   - Is it true (in the relevant sense)?\n   - Is it useful (generative, reframing, clarifying, or actionable)?\n\n4. **Keep probing** until you find material that passes the floor criteria.\n\n## Exit Triggers\n\nThe brainstorm session ends when:\n\n1. **User explicitly requests to move on** — natural language like \"I think I've got something\" or \"let's develop this one\"\n\n2. **System detects excitement signals** and offers to advance:\n   - Increased specificity (\"Oh, and another thing about that...\")\n   - Energy language (\"Yes!\", \"Exactly\", \"That's it\")\n   - Unprompted elaboration (user keeps building without being asked)\n   - Returning to an idea after exploring others\n\nWhen you detect excitement, offer: \"It sounds like this one is resonating—want to develop it further?\"\n\n## Transition\n\nWhen exiting brainstorm, transition to **Interview** to extract what's needed for an outline.\n\n## Lessons\n\n[Skill-specific lessons will be added here as they're captured]\n"
}