A professional MCP (Model Context Protocol) skill package for quantitative trading strategy research and backtesting.
- 📊 Data Access: Get stock historical data and stock lists
- 🧮 Factor Calculation: Calculate technical indicators (RSI, MACD, Momentum, etc.)
- 🧪 Backtesting: Run strategy backtests with realistic assumptions
- 📈 Strategy Evaluation: Multi-dimensional strategy performance evaluation
- 🔔 Real-time Monitoring: Stock price monitoring with alerts
-
Clone the repository
-
Install dependencies:
pip install numpy pandas
-
Configure database path in
config/config.py
Add to your MCP server configuration:
{
"mcpServers": {
"quantitative-strategy": {
"command": "python",
"args": ["src/mcp_tools.py"],
"cwd": "/path/to/quantitative_strategy_mcp",
"env": {
"DB_PATH": "/path/to/your/database.db"
}
}
}
}from src.mcp_tools import call_mcp_tool
# Get stock data
result = call_mcp_tool(
'get_stock_data',
ts_code='000001.SZ',
start_date='20240101',
end_date='20241231'
)
# Calculate factor
factor = call_mcp_tool(
'calculate_factor',
data=result['data'],
factor_type='rsi',
params={'period': 14}
)
# Run backtest
backtest = call_mcp_tool(
'run_backtest',
ts_code='000001.SZ',
start_date='20240101',
end_date='20241231',
strategy_type='momentum'
)- get_stock_data - Get stock historical data
- get_stock_list - Get stock list
- calculate_factor - Calculate quantitative factors
- run_backtest - Run strategy backtest
- evaluate_strategy - Evaluate strategy performance
- monitor_stocks - Monitor stock prices
See docs/QUANT_MCP_SKILL.md for detailed documentation.
Check examples/ directory for usage examples.
MIT License
马上有钱 (Robot)
1.0.0 (2026-03-15)
Contributions are welcome! Please see CONTRIBUTING.md for details.
This project is licensed under the MIT License - see the LICENSE file for details.
- Built with inspiration from ClawHub FinStep MCP
- Thanks to the quantitative trading community
- Issues: GitHub Issues
- Discussions: GitHub Discussions
Made with ❤️ by 马上有钱 (Robot)