VeriTrade

Autonomous Trading System Verification Framework

VeriTrade - AI-Powered Trading System Verification Framework

🎯 Mission

VeriTrade is a production-ready verification framework for algorithmic trading systems (Expert Advisors) on MetaTrader 5. The system autonomously verifies that a trading strategy executes correctly, safely, and profitably before deployment.

Verify before you trade - Never deploy an untested EA again.

✨ Features

• 🔍 Static Analysis - Code review and compliance checking
• 🏃 Runtime Testing - Execute EA and observe behavior in real-time
• ⚔️ Adversarial Testing - Edge cases and attack scenarios
• 🔄 Recovery Testing - Restart and recovery scenarios
• 📊 Evidence Collection - Screenshots, logs, and data capture
• 📈 Backtesting - Historical performance validation
• 🎯 Decision Engine - GO/NO-GO/NOT-PROVEN verdicts
• 🔒 Secure API - API key authentication and rate limiting
• 📝 Comprehensive Reports - HTML/JSON/Markdown output

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                    VeriTrade Framework                      │
├─────────────────────────────────────────────────────────────┤
│  ┌──────────────────────────────────────────────────────┐   │
│  │          1. MCP Server Layer                         │   │
│  │     - MT5 Control with Error Handling                │   │
│  │     - Tool Definitions                              │   │
│  │     - API Security                                  │   │
│  └──────────────────────────────────────────────────────┘   │
│                         │                                   │
│  ┌──────────────────────────────────────────────────────┐   │
│  │          2. Verification Controller                  │   │
│  │     - Static Analysis                               │   │
│  │     - Runtime Testing                               │   │
│  │     - Backtesting                                   │   │
│  │     - Evidence Collection                           │   │
│  └──────────────────────────────────────────────────────┘   │
│                         │                                   │
│  ┌──────────────────────────────────────────────────────┐   │
│  │          3. Reporting & Decision Engine              │   │
│  │     - Test Results                                  │   │
│  │     - Risk Assessment                               │   │
│  │     - GO/NO-GO/NOT-PROVEN Decision                  │   │
│  └──────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────┘

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/yourusername/veritrade.git
cd veritrade

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Install package
pip install -e .

Configuration

# Copy environment example
cp .env.example .env

# Edit with your MT5 credentials
nano .env

Usage

Start MCP Server

python scripts/run_server.py --port 8080 --host 0.0.0.0

Run Verification

python scripts/run_verification.py --ea GridEA_PathB.mq5 --symbol EURUSD --timeframe H1

Generate Report

python scripts/generate_report.py --run-id RUN-20260828-001 --format html

Programmatic Usage

from veritrade import VeriTrade

# Initialize
veritrade = VeriTrade()

# Run verification
result = veritrade.verify(
    ea_file="GridEA_PathB.mq5",
    symbol="EURUSD",
    timeframe="H1",
    test_types=["static", "runtime", "adversarial"]
)

# Check result
if result.verdict == "GO":
    print("✅ Ready for production deployment")
elif result.verdict == "NO-GO":
    print("❌ Needs fixes before deployment")
else:
    print("⚠️ More testing required")

📁 Project Structure

VeriTrade/
├── src/
│   ├── mcp_server/         # MCP server implementation
│   ├── controllers/        # MT5 and verification controllers
│   ├── core/              # Core utilities (error handling, logging, config)
│   ├── models/            # Data models and schemas
│   ├── tests/             # Test implementations
│   └── utils/             # Utility functions
├── scripts/               # CLI scripts
├── configs/               # Configuration files
├── tests/                 # Unit and integration tests
├── docs/                  # Documentation
├── examples/              # Example EAs and notebooks
└── .claude/commands/      # Claude agent commands

🧪 Testing

# Run unit tests
pytest tests/unit/

# Run integration tests
pytest tests/integration/

# Run with coverage
pytest --cov=src tests/

📊 Success Criteria

The system is production-ready when:

• ✅ All 6 phases implemented
• ✅ 80%+ code coverage
• ✅ End-to-end workflows work
• ✅ Complete documentation
• ✅ Security audit passed
• ✅ Verification < 30 minutes
• ✅ 99.9% uptime in testing
• ✅ All error scenarios handled
• ✅ Evidence collection working
• ✅ Accurate GO/NO-GO decisions
• ✅ Docker/PyPI deployment ready

📖 Documentation

• Architecture Guide
• API Reference
• User Guide

⚠️ Important Warnings

1. ❌ Never trade on live accounts during verification
2. ❌ Never skip error handling for speed
3. ❌ Never ignore evidence collection
4. ❌ Never assume tests pass without verification
5. ❌ Never deploy without GO decision
6. ❌ Never modify production EA during verification
7. ❌ Never use hardcoded credentials

🤝 Contributing

Contributions are welcome! Please read our contributing guidelines before submitting PRs.

📄 License

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

🙏 Acknowledgments

• MetaTrader5 team for the excellent Python integration
• MCP protocol contributors
• Trading community for best practices and insights

---

Remember: Verify before you trade! 🎯