Mantle AI Trader

AI-Powered Fundamental News-Based Trading Bot

Built for Mantle Turing Test Hackathon

🚀 Features

Core Trading Capabilities

• AI Signal Generation: Advanced signal generation powered by AI with multi-factor analysis
• Fundamental News Analysis: Real-time news aggregation from multiple sources with sentiment analysis
• Technical Analysis: Comprehensive technical indicators (RSI, MACD, SMA, EMA) and pattern recognition
• Risk Assessment: Intelligent risk scoring and position sizing recommendations

Trading Modes

• Demo/Paper Trading: Test strategies without real money
• Manual Mode: Execute signals manually with full control
• Auto-Trading: Automated signal execution (configurable)

Analysis Tools

• Backtesting Engine: Test strategies on historical data with performance metrics
• Signal Rating System: Track and rate signal performance
• Portfolio Analytics: Real-time P&L tracking and portfolio visualization

Integration

• Bybit API: Full integration with Bybit exchange (spot and futures)
• WebSocket Real-time Updates: Live price feeds and portfolio sync
• VectorDB: Semantic search for news and analysis

📊 Architecture

mantle-ai-trader/
├── src/
│   ├── app/                    # Next.js App Router
│   │   ├── api/                # API Routes
│   │   │   └── trading/        # Trading endpoints
│   │   └── page.tsx            # Main Dashboard
│   ├── lib/
│   │   ├── trading/
│   │   │   ├── core/           # Core types & Bybit client
│   │   │   ├── signals/        # Signal generation engine
│   │   │   ├── news/           # News aggregation
│   │   │   ├── backtest/       # Backtesting engine
│   │   │   └── demo/           # Paper trading
│   │   └── vector/             # VectorDB integration
│   └── components/ui/          # shadcn/ui components
├── mini-services/
│   └── trading-service/        # WebSocket service
├── prisma/
│   └── schema.prisma           # Database schema
└── tests/                      # Test files

🛠 Installation

Prerequisites

• Node.js 18+ or Bun
• SQLite (included)
• Bybit API keys (optional for live trading)

Setup

# Clone the repository
git clone https://github.com/yourusername/mantle-ai-trader.git
cd mantle-ai-trader

# Install dependencies
bun install

# Initialize database
bun run db:push

# Start development server
bun run dev

# Start WebSocket service (separate terminal)
bun run trading-service

📈 Usage

Dashboard

Access the dashboard at http://localhost:3000

1. Signals Tab: Generate AI signals for any supported trading pair
2. Positions Tab: View open positions and portfolio allocation
3. Backtest Tab: Run strategy backtests on historical data
4. News Tab: Browse latest market news with sentiment analysis

API Endpoints

Signals

# Generate signal
POST /api/trading/signals
Body: { "symbol": "BTCUSDT", "timeframe": "1h", "demo": true }

# Get signals
GET /api/trading/signals?symbol=BTCUSDT&limit=50

Demo Trading

# Get portfolio
GET /api/trading/demo?action=portfolio

# Place order
POST /api/trading/demo
Body: { "action": "place_order", "symbol": "BTCUSDT", "side": "BUY", "quantity": 0.01, "type": "MARKET" }

# Close position
POST /api/trading/demo
Body: { "action": "close_position", "symbol": "BTCUSDT" }

Backtest

# Run backtest
POST /api/trading/backtest
Body: {
  "symbol": "BTCUSDT",
  "startDate": "2024-01-01",
  "endDate": "2024-06-01",
  "initialCapital": 10000
}

News

# Get news
GET /api/trading/news?symbol=BTC&limit=20

# Get sentiment
GET /api/trading/news/sentiment?symbol=BTC

🔧 Configuration

Environment Variables

Create a .env file:

# Bybit API (optional - for live trading)
BYBIT_API_KEY=your_api_key
BYBIT_API_SECRET=your_api_secret
BYBIT_TESTNET=true

# News APIs (optional)
CRYPTOPANIC_API_KEY=your_key
CRYPTOCOMPARE_API_KEY=your_key

# ChromaDB (optional - for vector search)
CHROMADB_URL=http://localhost:8000

# Database
DATABASE_URL="file:./prisma/data/mantle-trader.db"

Risk Settings

Configure in UserSettings table or via API:

• Risk Level: CONSERVATIVE, MODERATE, AGGRESSIVE
• Max Position Size: Default $1,000
• Max Leverage: Default 5x
• Auto Trading: Enable/disable automatic execution

📊 Signal Analysis

Each signal includes:

Technical Analysis

• Trend direction and strength
• Support and resistance levels
• RSI, MACD, Moving Averages
• Candlestick patterns (Doji, Hammer, Engulfing, etc.)

Fundamental Analysis

• News impact score
• Market events summary
• Economic factors

Sentiment Analysis

• Overall sentiment score (-1 to 1)
• Sentiment label (Bullish/Bearish/Neutral)
• Key topics and trending keywords

Risk Assessment

• Risk score and level
• Suggested stop-loss and take-profit
• Position sizing recommendations

🧪 Testing

# Run unit tests
bun test tests/unit/

# Run integration tests
bun test tests/integration/

# Run all tests
bun test

📝 Documentation

• API Documentation
• Architecture Guide
• Contributing Guide

🔒 Security

• API keys are stored encrypted in the database
• All exchange communications use HTTPS
• WebSocket connections support authentication
• Rate limiting on all API endpoints

📜 License

MIT License - See LICENSE for details.

🏆 Mantle Turing Test Hackathon

This project is built for the Mantle Turing Test Hackathon:

• Track: AI Trading
• Prize Pool: $120,000 cash + $103,000 API credits
• Registration: mantle.to/Hackathon

Competition Tracks

• ✅ AI Trading - Trading bots, strategy automation, Bybit API
• ✅ AI Alpha & Data - Onchain analytics, anomaly detection

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Open a Pull Request

📞 Support

• GitHub Issues: Report a bug
• Telegram: Mantle Hackathon Chat

---

for Mantle Turing Test Hackathon

🌐 Website • 📦 Repository • 💬 Telegram