⚡ EnVisor AI — Electricity Audit App

AI-powered household electricity audit tool. Users scan their appliances via photo, and the app estimates monthly electricity costs with full analysis.

---

📁 Project Structure

envisor-ai/
├── backend/
│   ├── server.js                  # Express entry point
│   ├── routes/
│   │   ├── analyze.js             # POST /api/analyze/device|bill|summary
│   │   └── report.js              # POST /api/report/generate
│   ├── controllers/
│   │   ├── analyzeController.js   # Image AI analysis + summary logic
│   │   └── reportController.js    # Email report builder
│   ├── services/
│   │   └── aiService.js           # Anthropic Claude Vision API calls
│   ├── middleware/
│   │   └── upload.js              # Multer file upload config
│   └── config/
│       └── electricity.js         # PLN tariff rates & kWh calculations
├── frontend/
│   └── public/
│       ├── index.html             # Main app HTML
│       ├── css/
│       │   └── style.css          # All styles
│       └── js/
│           └── app.js             # Frontend logic (state, UI, API calls)
├── .env.example
├── .gitignore
├── package.json
└── README.md

---

🚀 Getting Started

1. Install dependencies

npm install

2. Configure environment

cp .env.example .env

Edit .env and set your Anthropic API key:

ANTHROPIC_API_KEY=sk-ant-...
PORT=3000

3. Run in development

npm run dev

4. Run in production

npm start

Open http://localhost:3000

---

🔌 API Endpoints

Method	Endpoint	Description
POST	/api/analyze/device	Analyze appliance photo → returns name, watts, cost
POST	/api/analyze/bill	Extract amount from PLN bill photo
POST	/api/analyze/summary	Full analysis from all rooms data
POST	/api/report/generate	Generate plain-text email report
GET	/api/health	Health check

POST /api/analyze/device

• Body: multipart/form-data with image (file) and va (number, optional)
• Response: { success, device: { name, watts, dailyHours, emoji, kwh, costPerMonth, costFormatted } }

POST /api/analyze/summary

• Body: application/json

{
  "rooms": [{ "id": "r1", "n": "Ruang Tamu", "i": "🛋️", "devs": [{ "n": "AC", "w": 900, "h": 8 }] }],
  "plnVa": 1300,
  "jumlahOrang": 3,
  "actualBill": 450000
}

---

🛠 Tech Stack

• Backend: Node.js + Express
• AI: Anthropic Claude Vision API (claude-opus-4-5)
• File uploads: Multer (memory storage)
• Frontend: Vanilla JS + CSS (no framework)

---

📝 Notes

• Photos are processed in-memory and never saved to disk
• Without ANTHROPIC_API_KEY, the frontend falls back to mock device data
• The frontend app.js has an API_URL constant — leave it empty ('') to use the same-origin Express server, or set it to your deployed URL (e.g. Railway)