New Idea

Capture a new idea while the context is fresh.

Instructions

1. Ask "What's the idea?"

   "What's the idea? Just dump what you're thinking — a sentence, a paragraph, a link. I'll sort it out."

   STOP and wait.

2. Extract a slug and one-liner from their response
   • Derive a short kebab-case slug for the directory name (e.g., book-to-audio, freight-ai-marketplace)
   • Write a one-liner summary of the idea
   • If anything is unclear, confirm the name before creating. Otherwise just go.

3. Scan conversation for context Look back through the current conversation for anything relevant to this idea:

   • Links discussed (repos, articles, competitor sites)
   • Technical details or constraints mentioned
   • Inspiration or prior art referenced
   • Problems or opportunities identified

   Capture this as notes/initial-context.md — a raw dump of everything relevant from the conversation, with sources.

4. Create the directory structure

   ideas/[idea-name]/
   ├── README.md              ← Overview, status, gut-check
   ├── notes/
   │   └── initial-context.md ← Auto-captured from conversation
   ├── research/              ← Market research outputs
   ├── competitors/           ← Competitor profiles and analysis
   ├── docs/                  ← Uploaded documents, pitch decks, financials
   └── notes/                 ← Meeting notes, strategy session outputs

5. Create README.md with this template:

   # [Idea Name]
   
   **Status**: Evaluating
   **Created**: [date]
   **Tags**: [suggest 2-3 tags like #saas, #side-project, #tool, #content, #marketplace, #ai, #b2b, #b2c, #dev-tool]
   **Go/No-Go**: Pending
   
   ## One-liner
   [One sentence summary derived from their description]
   
   ## Why this is interesting
   - [Bullet 1]
   - [Bullet 2]
   
   ## What could kill it
   - [Bullet 1]
   - [Bullet 2]
   
   ## Key Links
   | Resource | URL |
   |----------|-----|
   | [any links from conversation] | |
   
   ## Key People
   | Name | Role | Notes |
   |------|------|-------|
   | | | |
   
   ## Evaluation Progress
   - [ ] Opportunity brief captured
   - [ ] Market research complete
   - [ ] Competitor landscape mapped
   - [ ] Strategy session done
   - [ ] Due diligence complete
   - [ ] Viability assessment scored
   - [ ] Go/No-Go decision made
   
   ## Decision Log
   | Date | Decision | Rationale |
   |------|----------|-----------|
   | | | |

   Fill in the one-liner and key links from conversation context.

6. Ask for the gut-check

   "I've pre-filled what I could from our conversation. Two quick questions:

   Why is this interesting? What's the bet — why could this work?

   What could kill it? What's the biggest risk or reason this doesn't work?"

   STOP and wait. Fill in the gut-check sections with their answers.

7. Set this as the active idea Create/update .claude/active-opportunity.md:

   # Active Idea
   Currently evaluating: [Idea Name]
   Path: ideas/[idea-name]/

8. Suggest next step

   "You're set up. Next steps when you're ready:

   • /opportunity-brief — flesh out the full picture (model, people, risks)
   • /market-research — size the market and find demand signals
   • /competitor-landscape — see who else is playing here

   Or just keep chatting about it — I'll remember the context."