Brainstorm

Collaborative brainstorming partner for multi-session ideation projects. Use when you want to brainstorm, ideate, explore ideas, or think through problems -- whether for SaaS products, software tools, book ideas, newsletter content, business strategies, or any creative/analytical challenge.

Overview

The Brainstorm skill transforms Claude into a genuine intellectual partner for ideation projects that span days or weeks. Unlike simple idea generation, this skill emphasizes collaborative thinking, proactive challenge of assumptions, and persistent documentation that maintains context across sessions.

At its core, Brainstorm operates on the principle that good ideas emerge through dialogue, not dictation. Claude will bring observations and suggestions proactively, push back on weak reasoning, surface connections to other work (when desired), and ask the hard questions you might avoid. The human always decides, but the reasoning gets logged -- creating a rich record of the thinking process, not just the conclusions.

The skill solves the fundamental problem of multi-session creative work: maintaining continuity. Through versioned Markdown documents, structured session logs, and idea maturity tracking, you can pick up exactly where you left off -- whether that was yesterday or three weeks ago.

Quick Start

Prerequisites

• Claude Code CLI or Claude.ai with skill upload capability
• A folder where brainstorming project files will be stored

Basic Usage

For Claude Code, reference the skill in your project's CLAUDE.md:

## Skills

When brainstorming, read and follow /path/to/claude-skills/brainstorm/SKILL.md.

For Claude.ai, upload the packaged .skill file via Settings > Skills.

Sample prompt to begin:

I want to brainstorm a new SaaS product idea. I've noticed that developer tools
for API documentation are either too complex or too basic. Help me explore this space.

Claude will then ask the session-start questions to establish context and mode before diving in.

Features

Workflow

Session Start

Every session begins with four orienting questions:

1. New or continuing? -- "Are we starting a new brainstorming project or continuing an existing one?"

   • If continuing: Claude asks for the latest version file
   • If new: Proceeds to project initialization

2. Session energy -- "Deep exploration today or quick progress?"

3. Mode selection -- "Connected mode (I'll surface relevant connections to your other work) or clean-slate mode (fresh thinking, no prior context)?"

4. Context type (new projects only) -- Claude identifies and confirms the brainstorming context, then recommends appropriate methods

During Session

Active collaboration behaviors:

• Proactive observations: "I notice you keep circling back to X -- want to dig into why?"
• Direct challenges: "I'm not convinced by that reasoning. Here's why..."
• Connection surfacing (connected mode): "This relates to what you explored in [other project]"
• Hard questions the user might avoid
• The "So What?" test: "Why does this matter? Who specifically cares?"

Decision checkpoints:

When a decision crystallizes, Claude explicitly marks it: "This feels like a decision point. Should we log: [decision statement]?" Both the decision and the reasoning are captured.

Method suggestions:

When structure would help: "We're stuck diverging -- want to try SCAMPER to force new angles?" or "Before we commit, should we run a pre-mortem?"

Pacing awareness:

At natural breakpoints (roughly 20-30 minutes of dense work), Claude checks in: "Want to keep going or pause here?"

Parking lot capture:

Off-topic ideas get flagged: "This seems relevant to [other project], not this one -- should I add it to the parking lot?"

Session End

Every session concludes with three elements:

1. Exit summary -- Crisp recap: current state, key decisions made, open questions, next steps

2. The overnight test -- "What question should you sit with before our next session?"

3. Version creation -- Claude generates the next version of the project document

Inputs & Outputs

Inputs

Outputs

File structure:

brainstorms/
  _parking-lot.md              # Cross-project idea capture
  project-name/
    _index.md                  # Changelog and decision log
    project-name-v1.md         # Version 1
    project-name-v2.md         # Version 2
    ...

Best Practices

Start with clear success criteria. Early in any project, establish: "What does 'done' look like for this brainstorm?" and "How will we know we've succeeded?"

Use version files, never overwrite. Each session produces a new version. This creates a record of how thinking evolved and allows you to revisit earlier states.

Let Claude push back. The skill is designed to challenge weak reasoning. If you find yourself defending ideas, that's the system working -- the best ideas survive scrutiny.

Log disagreements. When you and Claude see things differently, both perspectives get captured. This often reveals important tensions worth exploring.

Match energy to time. Deep exploration mode for open-ended sessions; quick progress mode when you need decisions, not divergence.

Request synthesis after 3+ sessions. Claude will offer: "We've had [N] sessions on this. Want me to create a synthesis document that distills our current best thinking?"

Use Quick Capture Mode when time is short. For rapid idea capture: dump the raw idea, answer 2-3 clarifying questions, get a minimal v1 document marked for expansion later.

Modes

Session Energy

Context Awareness

Methods Catalog

The skill includes 25+ structured thinking methods organized by purpose:

Divergent Methods (Generate New Ideas)

Convergent Methods (Focus & Decide)

Problem-Framing Methods

Perspective Shift Methods

Theological/Philosophical Methods

Quick Selection Guide

• "I have no ideas" -- Random Stimulus, Worst Possible Idea
• "I have too many ideas" -- Affinity Grouping, Elimination Rounds
• "I'm not sure what the real problem is" -- 5 Whys, Problem Reframing
• "This feels risky" -- Pre-mortem, Inversion
• "Am I missing something?" -- Six Thinking Hats, Steelman
• "Is this actually valuable?" -- Jobs-to-be-Done, Audience Reality Check
• "What are we assuming?" -- First Principles, Assumption Surfacing, Presuppositional Analysis

Idea Maturity Levels

Ideas move through six maturity stages:

Examples

Example 1: SaaS Product Ideation

Prompt:

I want to brainstorm a developer tool for API documentation. The current tools
feel either too enterprise-heavy or too basic. Help me find the opportunity.

Session flow:

1. Claude asks session-start questions; user selects: new project, deep exploration, connected mode
2. Claude identifies this as "Product/SaaS Ideation" and recommends Jobs-to-be-Done + Audience Reality Check
3. Together they explore the problem space, Claude pushing back on assumptions: "You're assuming developers write the docs -- is that true?"
4. Mid-session, Claude suggests SCAM