Skip to content

brainstorm.md

Latest commit

 

History

History
214 lines (162 loc) · 7.51 KB

brainstorm.md

File metadata and controls

214 lines (162 loc) · 7.51 KB
Raw
name brainstorm
model reasoning
description Run a structured brainstorming session to generate, filter, and expand ideas for features, solutions, or designs
usage /brainstorm [topic] [--mode <diverge|converge|both>]

/brainstorm

Run a structured brainstorming session that generates diverse ideas, filters them by impact and feasibility, and expands the top picks into actionable proposals.

Usage

/brainstorm [topic] [--mode <diverge|converge|both>]

Arguments:

  • topic — The problem, opportunity, or question to brainstorm around
  • --mode <mode> — Session mode: diverge (generate ideas only), converge (filter and rank existing ideas), both (full session, default)

Examples

/brainstorm "How can we reduce onboarding time from 10 minutes to under 2?"
/brainstorm "New features for Q3 roadmap" --mode diverge
/brainstorm "Architecture for real-time collaboration" --mode both
/brainstorm "Monetization strategies for free tier users" --mode converge
/brainstorm "Reduce API response time below 100ms"

When to Use

  • Starting a new feature and need to explore approaches before committing
  • Stuck on a problem and need fresh angles
  • Planning a roadmap and need to generate feature candidates
  • Evaluating tradeoffs between multiple solutions
  • Exploring how to 10x an existing experience
  • Before any significant architecture or product decision

What It Does

  1. Defines the problem or opportunity with clear boundaries
  2. Diverges to generate 10+ ideas using multiple creative lenses
  3. Converges to filter and rank ideas by impact and effort
  4. Expands the top 3 ideas into structured proposals with user stories, requirements, and risks
  5. Presents ranked recommendations with clear rationale

Implementation Steps

Use TodoWrite to track progress through each phase.

Phase 1: Define the Problem

Before generating ideas, frame the problem clearly:

Element Question
Problem statement What exactly are we trying to solve or improve?
Current state What exists today? What's working? What's not?
Success criteria How will we know the solution worked?
Constraints What are the non-negotiables? (timeline, budget, tech stack, team size)
Scope What's in bounds? What's explicitly out of bounds?

Write a one-sentence problem statement in this format:

"How might we [desired outcome] for [target user] given [key constraint]?"

Phase 2: Diverge — Generate Ideas

Generate at least 10 ideas. Quantity matters more than quality at this stage. Use multiple lenses to force diversity:

Lens Prompt
User perspective What would the user wish this could do?
10x thinking What if we made this 10x better, not 10% better?
Inversion What's the opposite approach? What if we removed this entirely?
Analogy How do other industries/products solve a similar problem?
Constraint removal If we had unlimited time/money/engineers, what would we build?
Simplification What's the simplest possible solution that still works?
Composition Can we combine two existing features to create something new?
Automation What's currently manual that could be automated?
Edge cases What would power users want? What about complete beginners?
Competitor What would [top competitor] do? What would they never do?

Rules for divergence:

  • No judging ideas during this phase
  • Write each idea in one sentence
  • Include wild ideas — they often spark practical ones
  • Build on previous ideas ("Yes, and…")
  • Aim for at least 10, push to 15+ if the topic is rich

Phase 3: Converge — Filter and Rank

Score each idea on two dimensions:

Dimension Scale Criteria
Impact 1–5 How much does this move the needle on the success criteria?
Effort 1–5 How much time, complexity, and risk does this involve? (1 = low effort)

Calculate priority score: Impact × (6 - Effort) — this naturally surfaces high-impact, low-effort ideas.

Create a 2×2 matrix:

         High Impact
              │
   Quick Wins │ Big Bets
   (do first) │ (plan carefully)
──────────────┼──────────────
   Fill-ins   │ Money Pits
   (do if easy)│ (avoid)
              │
         Low Impact
    Low Effort ──── High Effort

Select the top 3 ideas (prioritizing Quick Wins, then Big Bets).

Phase 4: Expand Top Ideas

For each of the top 3 ideas, develop a structured proposal:

User Story:

As a [type of user],
I want to [action/capability],
so that [benefit/outcome].

Key Requirements:

  • 3–5 must-have requirements
  • 2–3 nice-to-have requirements

Technical Considerations:

  • Architecture implications
  • Dependencies on existing systems
  • Data model changes
  • API changes
  • Third-party integrations needed

Risks and Tradeoffs:

Risk Likelihood Impact Mitigation
[risk description] Low/Med/High Low/Med/High [how to reduce]

Estimated Effort:

  • T-shirt size: XS / S / M / L / XL
  • Key milestones and rough timeline

Phase 5: Present Recommendations

Rank the top 3 ideas with a clear recommendation:

Rank Idea Impact Effort Recommendation
1 [name] [score] [score] [why this is #1]
2 [name] [score] [score] [why this is #2]
3 [name] [score] [score] [why this is #3]

Include a "Start with…" recommendation — the single next action to take.

NEVER Do

Rule Reason
NEVER judge ideas during divergence Premature filtering kills creative thinking and reduces idea diversity
NEVER stop at the first good idea The best idea is rarely the first one. Push for volume.
NEVER skip the problem definition Brainstorming without a clear problem produces unfocused, unusable ideas
NEVER present ideas without ranking Unranked lists create decision paralysis. Always prioritize.
NEVER ignore feasibility Even in divergent thinking, note obviously impossible ideas so convergence is efficient
NEVER expand more than 3 ideas Deep analysis of too many ideas wastes effort. Go deep on the best few.

Output

Brainstorm: [topic]
===================
Problem: [one-sentence problem statement]
Mode:    [diverge | converge | both]

All Ideas (ranked):
  1. [idea] — Impact: X, Effort: X, Score: X
  2. [idea] — Impact: X, Effort: X, Score: X
  ...

Top 3 Expanded:

#1: [Idea Name]
  User Story: As a [user], I want [capability], so that [benefit]
  Requirements: [list]
  Technical: [considerations]
  Risks: [table]
  Effort: [t-shirt size]

#2: [Idea Name]
  ...

#3: [Idea Name]
  ...

Recommendation: Start with [#1] because [rationale].
Next action: [specific first step].

Related

  • Skill: brainstorming — structured ideation frameworks
  • Skill: game-changing-features — finding 10x product opportunities
  • Skill: decision-frameworks — structured decision-making tools
  • Agent: development — development agent
  • Command: /new-feature — take a brainstormed idea into implementation
  • Command: /write-content — write up a brainstorm outcome as a proposal document