Trading-MCP by Bwizrd - MCP Server

📊 Trading MCP Server

A comprehensive Model Context Protocol (MCP) server for trading strategy analysis and chart generation. This project provides advanced trading analytics, backtesting capabilities, and professional-grade chart visualization for forex trading strategies.

🚀 Features

📈 Comprehensive Trading Charts

Unified Chart System: Single chart type with ALL advanced features
1-Minute Precision: Uses 1-minute data for accurate entry/exit timing
Professional Visualization: Interactive Plotly charts with multiple sections

🎯 Chart Components

Candlestick Price Action: OHLC data with green/red coloring
VWAP Overlay: Volume-weighted average price line (calculated from 1m data)
Entry/Exit Markers: Numbered circles and X marks with color coding
Connection Lines: Colored lines linking entries to exits (Green=profit, Red=loss, Orange=break-even)
Cumulative P&L: Real-time running total of strategy performance
Trade Summary Table: Complete trade details with color-coded results
Total P&L Display: Prominent bottom display of net strategy performance

🔍 Advanced Analytics

VWAP Strategy: Entry signals based on price vs VWAP relationship
Precise Backtesting: Minute-level accuracy for realistic results
Risk Management: Configurable stop-loss and take-profit levels
Performance Metrics: Win rate, total pips, trade-by-trade analysis

📁 Project Structure

Trading-MCP/
├── mcp_servers/                    # All MCP server implementations
│   ├── charts/                    # Chart generation servers
│   │   └── trading_charts_mcp.py # Main chart server
│   └── strategies/                # Trading strategy servers
│       └── vwap_strategy/         # VWAP strategy implementations
│           ├── core.py           # Core strategy (mock data)
│           ├── ctrader.py        # cTrader integration
│           └── influxdb.py       # InfluxDB + cTrader integration
├── shared/                        # Shared modules and utilities
│   ├── models/                   # Common Pydantic models
│   └── utils/                    # Utility functions
├── config/                        # Configuration files
│   └── settings.py              # Application settings
├── data/                         # Generated data and outputs
│   ├── charts/                  # Generated chart outputs
│   └── optimization_results/    # Backtest and optimization data
├── requirements.txt              # Python dependencies
├── README.md                     # This file
├── ARCHITECTURE.md              # Technical architecture overview
├── QUICKSTART.md               # Quick setup guide
├── UNIFIED_CHARTS.md           # Chart system documentation
└── 1MINUTE_PRECISION_UPDATE.md # Latest precision improvements

🛠️ Installation

Prerequisites

# Python 3.10 or higher required
python --version

# Install required packages
pip install -r requirements.txt

Setup

Clone the repository:

git clone <repository-url>
cd Trading-MCP

Set up environment variables:

export CTRADER_API_URL="http://localhost:8000"
export CTRADER_API_USERNAME="your_username"
export CTRADER_API_PASSWORD="your_password"

🎮 Usage

Starting the MCP Server

python trading_charts_mcp.py

Creating Charts

Use the MCP client to request charts:

{
  "symbol": "EURUSD",
  "start_date": "2025-10-20",
  "end_date": "2025-10-25",
  "timeframe": "30m",
  "stop_loss_pips": 10,
  "take_profit_pips": 15
}

Output

Interactive HTML charts saved to charts/ directory
Complete trading analysis with all features included
Professional-grade visualization ready for sharing

📊 Chart Features

Upper Section: Price Chart with Trades

Candlestick price action
VWAP overlay line
Numbered entry markers (🟢 BUY, 🔴 SELL)
Numbered exit markers (✅ WIN, ❌ LOSS, 🟨 EOD)
Colored connection lines between entries and exits

Middle Section: Cumulative P&L

Running total of strategy performance
Visual trend analysis
Zero reference line

Bottom Section: Trade Summary

Complete trade details table
Color-coded pip results
Entry/exit prices and times
Win/Loss/EOD status

Total P&L Display

Prominent bottom summary
Color-coded total performance
Large, bold formatting

🔧 Configuration

Supported Symbols

Major forex pairs: EURUSD, GBPUSD, USDJPY, AUDUSD, etc.
Indices: GER40, UK100, US30

Timeframes

Chart display: 15m, 30m, 1h, 4h, 1d
Backtest precision: Always 1-minute for accuracy

Strategy Parameters

Configurable stop-loss (default: 10 pips)
Configurable take-profit (default: 15 pips)
Entry time: 8:30 AM (or closest available)

Integration with Claude

Claude Desktop Configuration

Locate your Claude Desktop config file:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
- Linux: ~/.config/Claude/claude_desktop_config.json
Add the MCP server to your config:

{
  "mcpServers": {
    "trading-charts": {
      "command": "python",
      "args": ["/full/path/to/Trading-MCP/mcp_servers/charts/trading_charts_mcp.py"]
    },
    "vwap-strategy-core": {
      "command": "python", 
      "args": ["/full/path/to/Trading-MCP/mcp_servers/strategies/vwap_strategy/core.py"]
    },
    "vwap-strategy-ctrader": {
      "command": "python", 
      "args": ["/full/path/to/Trading-MCP/mcp_servers/strategies/vwap_strategy/ctrader.py"]
    },
    "vwap-strategy-influxdb": {
      "command": "python", 
      "args": ["/full/path/to/Trading-MCP/mcp_servers/strategies/vwap_strategy/influxdb.py"]
    }
  }
}

Important: Replace /full/path/to/ with the actual absolute path where you saved the file.

Restart Claude Desktop

🏗️ Architecture

MCP Server Design

FastMCP: Modern Python MCP server framework
Async/Await: High-performance async operations
cTrader Integration: Direct API connectivity
Plotly Visualization: Professional interactive charts

Data Flow

1-Minute Data Fetch: High-resolution data from cTrader API
Chart Data Aggregation: User timeframe for visualization
Precise Backtesting: Minute-level entry/exit detection
Comprehensive Output: Multi-section interactive chart

🤝 Contributing

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

📄 License

This project is licensed under the MIT License - see the file for details.

🔗 Related Documentation

📞 Support

For questions or issues:

Check existing documentation
Review the code comments
Create an issue in the repository

🎯 Professional trading analysis with institutional-grade precision!

# Check Python version
python --version  # Should be 3.10+

# Check dependencies
pip list | grep -E "mcp|httpx|pydantic"

# Check for syntax errors
python -m py_compile trading_strategy_mcp.py

Claude Can't See the Server

Verify config file location and syntax
Use absolute paths (not ~/ or relative paths)
Restart Claude Desktop completely
Check Claude Desktop logs for errors

API Connection Issues

If using real API:

Verify API key is valid
Check API endpoint URLs
Test API independently with curl/Postman
Check for rate limiting

Security Notes

⚠️ Important Security Considerations:

API Keys: Store API keys in environment variables, not hardcoded:

import os
API_KEY = os.environ.get("TRADING_API_KEY")

Read-Only Mode: This server is designed for analysis only. For actual trade execution, additional safety measures are required.
Data Validation: All inputs are validated with Pydantic models to prevent injection attacks.

Next Steps

Connect your real data feed - Replace mock functions
Test thoroughly - Backtest on historical data
Add more strategies - Create new tools for different approaches
Add filters - Market conditions, news events, volatility
Build monitoring - Track live performance

Support

For questions about:

MCP Protocol: https://modelcontextprotocol.io
Claude Integration: https://docs.claude.com
This Implementation: Review the code comments

License

This is a demonstration project. Use at your own risk. Not financial advice.