nubiv/my-nome/brainstorm

Brainstorm

Quick start

When the user wants to brainstorm, run an interactive session:

1. Clarify the topic - Ask what they want to brainstorm about and any constraints
2. Choose approach - Pick the best framework for the topic (or go freeform)
3. Generate ideas - Produce a wide range of ideas, then cluster and refine
4. Converge - Help the user narrow down to actionable next steps

Frameworks

Use the framework that fits the topic. Default to freeform if none is a clear match.

First Principles Thinking

Break the problem down to fundamental truths, challenge assumptions, then rebuild:

1. State the problem clearly
2. List assumptions people typically make about it
3. Challenge each assumption - is it actually true? What evidence?
4. Identify fundamentals - what remains after removing assumptions?
5. Rebuild solutions from the ground up using only fundamentals

Best for: technical architecture, cost reduction, "why do we do it this way?" questions.

Six Thinking Hats

Examine the topic from six perspectives sequentially:

• White hat - Facts and data. What do we know? What's missing?
• Red hat - Feelings and intuition. Gut reactions without justification.
• Black hat - Caution. Risks, downsides, what could go wrong.
• Yellow hat - Optimism. Benefits, best-case outcomes, value.
• Green hat - Creativity. New ideas, alternatives, provocations.
• Blue hat - Process. Summary, next steps, decisions.

Best for: evaluating proposals, team decisions, balancing perspectives.

Freeform Diverge-Converge

1. Diverge - Generate as many ideas as possible without judgment
2. Cluster - Group related ideas into themes
3. Evaluate - Score each cluster on feasibility and impact
4. Converge - Select top ideas and define next steps

Best for: feature ideation, open-ended exploration, creative solutions.

Output format

Adapt output to the topic:

• Short list for quick ideation (5-10 bullet points)
• Categorized ideas with brief pros/cons for feature brainstorming
• Hierarchical breakdown for problem decomposition
• Decision matrix (table with criteria scores) when comparing options

Guidelines

• Quantity over quality during divergence - filter later
• Challenge the user's framing if it seems too narrow
• Suggest unexpected angles and cross-domain analogies
• Ask clarifying questions between phases rather than all upfront
• End every session with concrete next steps or decisions to make