MindSplit is an AI-powered decision analysis tool that visualises the logical and emotional paths of any meaningful decision you're facing — step by step, as an interactive node graph.

Describe your scenario. Watch it split into two paths. Choose your steps. See where each decision leads.

• Dual-path node graph — logical (Jekyll) vs emotional (Hyde) paths rendered as a horizontal Blender-style graph with animated SVG bezier connectors
• Step-by-step progression — reveal one column at a time; each choice shapes the path forward
• Coin-flip external circumstances — random real-world events that modify outcomes at each step
• Short-term & long-term impact scores — tracked cumulatively as you build your path
• Smart input validation — client-side heuristics + AI-powered classifier reject vague, trivial, or non-decision inputs before hitting the main API
• How-to guide — built-in modal walkthrough, auto-shown on first visit

git clone https://github.com/Hestone/MindSplit.git
cd mindsplit
npm install

Create a .env.local file in the project root:

GEMINI_API_KEY=your_gemini_api_key_here

Get a free API key at aistudio.google.com.

npm run dev

Open http://localhost:3000 in your browser.

app/
├── page.tsx              # Main UI — node graph, state, interaction logic
├── globals.css           # Canvas background, node animations, transition classes
├── layout.tsx            # Root layout
├── components/
│   ├── Header.tsx        # MindSplit gradient title
│   └── InputCard.tsx     # (reserved)
└── api/
    └── analyze/
        └── route.tsx     # POST /api/analyze — Gemini classifier + generation

1. Describe a real decision or dilemma you're facing
2. MindSplit validates your input (client-side heuristics → AI classifier)
3. Gemini generates 4 logical steps, 4 emotional steps, and 4 external circumstances
4. The node graph renders — choose Logical or Emotional at each step
5. A random circumstance event fires at each step, modifying your score
6. After 4 steps, your final impact scores and a summary are revealed

npm run dev      # Start development server
npm run build    # Production build
npm run start    # Start production server
npm run lint     # Run ESLint

MIT