Brainstorm and mind-map with AI
Skills used in this guide (3)
任何创造性工作之前都必须用它——创建功能、构建组件、添加功能或修改行为。先探索用户意图、需求与设计，再实现。
npx skills add obra/superpowers -g（占位）源数据描述为空
npx skills add nexu-io/open-design -gSVG 曲线连接、便利贴节点、白板 brainstorm 风格的光标交互。
npx skills add nexu-io/open-design -g
The worst brainstorming is "diverge and stop". The point of AI help: first force enough candidates (quantity before quality), then converge on 3 doable ones by "feasibility / impact" — not stop at "many ideas".
Who this is for / not for
For:
- Need ideas for features, campaigns, content
- Stuck at "can't think" or "too many to pick"
- Want diverge + converge two-phase output
Not for:
- Answer already set, just packaging (that's copy, not brainstorm)
- Serious feasibility论证 (brainstorm ≠ evaluation)
- Thinking alone, not letting AI try
Before you start
- Topic: the problem to brainstorm (specific beats vague).
- Constraints: budget, time, audience.
- (Optional) draft: let AI extend it.
Steps
1. Install the skill
brainstorming is a brainstorm skill. Copy its folder into your client's skills directory (same as any Agent Skill):
| Client | Directory |
|---|---|
| Claude Code | ~/.claude/skills/ |
| Codex | ~/.codex/skills/ |
| Cursor | ~/.cursor/skills/ |
| Trae | ~/.trae/skills/ |
| CodeBuddy | ~/.codebuddy/skills/ |
| Tongyi Lingma | ~/.lingma/skills/ |
2. Diverge, then converge
Around "how to lift new-user week-1 retention", first diverge 20+ ideas, no judging;
then by feasibility(h/m/l) × impact(h/m/l), pick the 3 to do first, each with one action.3. Land into steps
With many ideas, use frame-flowchart-sticky to lay candidates as sticky flow, see order and deps:
For each of the 3 ideas, break into first-step / blocker / next-step sticky structure,
mark what can parallel, what depends.4. You pick direction
AI gives candidates; you decide final direction; keep rejected ones noted for reuse.
Pitfalls
| Pit | Symptom | Fix |
|---|---|---|
| Diverge only | Many, no conclusion | Force 2-axis converge to 3 |
| Judge early | Quantity low | No judging first, filter later |
| Too vague | Can't land | Each with one action |
| No constraint | Unfeasible | Give budget/time/audience |
| Trust AI pick | Off intent | You decide final |
FAQ
Brainstorm with no topic?
Need a question first; more specific → sharper ideas.
How much diverge?
Quantity first, force 15-20 before filtering, no early stop.
Export mindmap?
Sticky/flow → Mermaid or text outline, then into a diagram tool.
Which for product?
Design → nexu brainstorming; general structured → obra's.
Convergence standard?
Feasibility × impact axes; prioritize high-high, then high-medium.
Related guides
Continue across these scenes:
相关源码
- obra/superpowers · brainstorming source
- nexu-io/open-design · brainstorming source
- nexu-io/open-design · frame-flowchart-sticky source
- Agent Skill open format spec
一句话结论
结论：脑暴先发散后收敛——逼出足量候选再按"可行性×影响"收到能动手的 3 个，不停在"想法很多"。
References
Related Scenes
Generate illustrations and marketing images with AI
The worst text-to-image is "one-line prompt, pure luck". The point: set style ref + subject + composition + ratio as structured prompt, then iterate small — not throw "draw something nice" and hope.
Generate icons and article illustrations with AI
The pain of illustration isn't "can't draw" — it's "inconsistent style, blurry when scaled, can't edit". The point of AI vector illustration: set style and size first, want vector not bitmap, so it scales sharp and the source stays editable.
Make podcast scripts and voiceover with AI
The pain of podcast/talking-head isn't "no voice" — it's "script reads like a paper, voice sounds robotic, nobody finishes". The point of AI: write the script as spoken human language, pick a natural voice, then decide on visuals — don't pile effects upfront.
Turn the architecture in your head into a diagram with AI
The pain of diagramming isn't "can't draw" — it's that edits are exhausting: change one box and you re-layout the whole thing. Generating descriptive, re-flowable diagrams turns "drawing" into "describing structure"; editing text is 10x faster than editing the picture.