🐅 Tiger MCP System
Professional-grade Model Context Protocol (MCP) server for Tiger Brokers API integration with Claude AI
A comprehensive MCP server system that enables seamless integration between Claude AI and Tiger Brokers trading platform, featuring real-time market data, algorithmic trading capabilities, and intelligent portfolio management.
✨ Key Features
🤖 AI-Native Trading
- Claude AI Integration: Direct MCP protocol support for conversational trading
- Natural Language Processing: Execute trades using natural language commands
- Intelligent Analysis: AI-powered market analysis and portfolio optimization
- Risk Management: Automated risk assessment and position sizing
📊 Real-Time Market Data
- Live Market Feeds: Real-time quotes, order books, and market depth
- Historical Analysis: Comprehensive historical data with technical indicators
- Market Scanning: Advanced screening and filtering capabilities
- Global Markets: Support for US, HK, SG markets and more
- Sub-second Execution: Optimized order routing and execution
- Portfolio Management: Real-time position tracking and P&L monitoring
- Risk Controls: Built-in risk management and compliance checks
- Scalable Architecture: UV workspace with modern async/await patterns
🛡️ Enterprise Security
- Encrypted Communication: TLS/SSL encryption for all data transmission
- Secure Authentication: Tiger Brokers OAuth2 with refresh token management
- Audit Logging: Comprehensive logging for compliance and debugging
- Production Ready: Docker containerization with health checks
🚀 Quick Start
Prerequisites
- Python 3.11+ with UV package manager
- Docker & Docker Compose for containerized deployment
- Tiger Brokers Account with API access (Apply here)
- Claude Desktop or Claude API access
⚡ 30-Second Setup
# 1. Clone and setup
git clone <your-repository-url>
cd tiger-mcp
cp .env.template .env # Configure your Tiger API credentials
# 2. Install dependencies
curl -LsSf https://astral.sh/uv/install.sh | sh # Install UV
uv sync # Install all workspace dependencies
# 3. Start with Docker
docker-compose up -d
# 4. Verify installation
curl http://localhost:8000/health
🔧 Configuration
Edit .env with your Tiger Brokers credentials:
# Tiger API Configuration
TIGER_CLIENT_ID=your_client_id_here
TIGER_PRIVATE_KEY=your_private_key_here
TIGER_ACCOUNT=your_account_id_here
TIGER_SANDBOX=true # Set to false for live trading
# Database (using Docker defaults)
DATABASE_URL=postgresql://tiger:tiger@localhost:5432/tiger_mcp
⚠️ Security Note: Never commit real credentials to version control. Use .env file for local development and secure environment variables for production.
🤖 Claude AI Integration
Tiger MCP supports both Claude Desktop and Claude Code with comprehensive trading capabilities.
Claude Desktop Configuration
Add Tiger MCP to your Claude Desktop configuration (~/.config/claude/config.json):
{
"mcpServers": {
"tiger-mcp": {
"command": "uv",
"args": ["run", "--package", "mcp-server", "python", "-m", "mcp_server.main"],
"cwd": "/path/to/tiger-mcp",
"env": {
"TIGER_CLIENT_ID": "your_client_id",
"TIGER_PRIVATE_KEY": "your_private_key",
"TIGER_ACCOUNT": "your_account_id"
}
}
}
}
Claude Code Integration
For Claude Code users, Tiger MCP provides enhanced development and trading workflows:
# Configure Claude Code MCP server
cp config/claude_mcp.json ~/.config/claude-code/mcp-servers/
# Start the MCP server for Claude Code
uv run --package mcp-server python -m mcp_server --claude-code-mode
# Use with Claude Code commands
claude code analyze-portfolio --account=all
claude code backtest-strategy --symbol=AAPL --days=30
claude code risk-assessment --portfolio
Claude Code Features:
- Code-Driven Trading: Generate and execute trading algorithms
- Portfolio Analysis: Automated portfolio analysis and reporting
- Strategy Development: Backtest and optimize trading strategies
- Risk Management: Code-based risk analysis and position sizing
- Multi-Account Operations: Manage multiple Tiger accounts programmatically
🎯 Example Conversations
General Trading (Claude Desktop & Claude Code):
💬 "Show me my portfolio performance this week"
💬 "What are the top gainers in tech stocks today?"
💬 "Buy 100 shares of AAPL when it drops below $180"
💬 "Analyze the risk of my current positions"
Multi-Account Trading:
💬 "Show portfolio for account account2_id"
💬 "Compare performance across all my accounts"
💬 "Switch to trading account account3_id and buy 100 TSLA"
💬 "What's the total portfolio value across all accounts?"
Claude Code Advanced Usage:
💬 "Generate a momentum trading strategy for NASDAQ stocks"
💬 "Backtest a mean reversion strategy on my current holdings"
💬 "Create a risk report for all accounts in PDF format"
💬 "Implement dollar-cost averaging for my portfolio"
| Tool | Description | Use Case |
|---|
get_account_info | Account details and balances | Portfolio overview |
get_portfolio | Holdings and P&L analysis | Performance tracking |
get_market_data | Real-time quotes and data | Market analysis |
place_order | Execute trading orders | Order management |
scan_market | Market screening tools | Stock discovery |
get_historical_data | Historical price data | Backtesting |
calculate_indicators | Technical analysis | Chart analysis |
📖 Documentation
🚀 Getting Started
🔧 Development
🚀 Deployment
🎯 Use Cases
💼 Portfolio Management
- Real-time portfolio tracking and performance analysis
- Automated rebalancing and risk management
- Multi-account portfolio consolidation
- Tax-loss harvesting optimization
📈 Algorithmic Trading
- Strategy backtesting with historical data
- Automated order execution based on technical indicators
- Custom alert systems for market opportunities
- High-frequency trading with sub-second latency
🔍 Market Research
- Advanced market scanning and stock screening
- Fundamental and technical analysis automation
- Sector rotation and correlation analysis
- Options strategy analysis and optimization
🤖 AI-Powered Insights
- Natural language queries for market data
- Automated report generation and analysis
- Risk assessment and scenario modeling
- Intelligent trade recommendations
🏗️ Architecture
Tiger MCP is built using a modern, scalable architecture:
- MCP Server: FastMCP-based server with async/await patterns
- Database: PostgreSQL with async SQLAlchemy for data persistence
- Caching: Redis for high-performance data caching
- Security: JWT authentication with encrypted credentials storage
- Monitoring: Comprehensive logging and health check endpoints
🤝 Contributing
We welcome contributions! Please see our Contributing Guidelines for details.
Development Process
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Code Standards
- Python: Black formatting, type hints, docstrings
- Testing: Minimum 80% test coverage required
- Documentation: Update docs for any new features
- Security: Follow security best practices
📜 License
This project is licensed under the MIT License - see the LICENSE file for details.
⚠️ Disclaimer
Important: This software is for educational and research purposes. Trading financial instruments involves substantial risk and may not be suitable for all investors. Past performance does not guarantee future results. Always consult with financial professionals before making investment decisions.
🆘 Support & Community
🙏 Acknowledgments
- Tiger Brokers for providing robust API services
- Anthropic for Claude AI and MCP protocol
- FastMCP for the excellent MCP framework
- UV for modern Python package management
⭐ Star this repository if you find it useful!
Built with ❤️ for the trading and AI community