RustTrade Agentic Bot 🦀

[!WARNING] This project is not production-ready and should not be used with real money. Using this trading bot involves significant financial risks. Use it only in simulation mode (Paper Trading) or for educational purposes.

A high-performance, multi-agent algorithmic trading system built in Rust. Capable of real-time market surveillance, trend analysis, and autonomous execution. Now featuring a full Native Desktop Interface.

📸 Screenshots

🚀 Key Features

🧠 Intelligent Agents

• Multi-Agent Architecture: 6 specialized agents (Sentinel, Scanner, Analyst, Risk Manager, Order Throttler, Executor).
• Regime Detection: Automatically detects Bull, Bear, Sideways, and Volatile market regimes.
• Dynamic Scanning: Real-time discovery of "Top Movers" and volatility opportunities (Stocks & Crypto).
• System Health Monitoring: Centralized ConnectionHealthService that tracks physical connectivity and broadcasts status to all agents for safe execution.
• Advanced Observability: Full tracing instrumentation, custom Prometheus metrics (win rate, drawdown, latency), and structured logging.

🖥️ Native User Interface (New)

• High-Performance Dashboard: Built with egui (0.31) for zero-latency monitoring.
• Real-Time Visuals: Live charts, interactive metric cards, and scrolling activity feed.
• Internationalization (i18n): Fully localized in English 🇬🇧 and French 🇫🇷.
• Manual Control: "Panic Button" and manual strategy overrides directly from the UI.

📈 Trading Strategies

• Standard: Dual SMA Crossover (Golden/Death Cross).
• Advanced: Triple Filter System (SMA Trend + RSI Momentum + MACD + ADX Trend Strength).
• Dynamic: Automatically switches strategies based on Market Regime (e.g., Trend Riding vs Mean Reversion).
• Statistical: Z-Score Mean Reversion & Statistical Momentum (Regression Slope).
• Machine Learning: Random Forest Classifier trained on technical features (MLStrategy).
• Trend Riding: Long-term trend following with trailing stops.
• Mean Reversion: Contrarian strategy for overbought/oversold conditions (BB + RSI).

🛡️ Advanced Risk Management

• Correlation Filters: Prevents exposure to highly correlated assets (Pearson matrix).
• Sector Exposure Limits: Caps risk per industry sector.
• Circuit Breakers: Daily Loss Limit, Max Drawdown Halt, Composite Risk Score.
• Persistence ("No Amnesia"): Retains critical risk state (HWM, Daily Loss) across restarts.
• Panic Mode: Blind emergency liquidation during data outages.
• PDT Protection: Pattern Day Trader safeguards for accounts < $25k.
• Singleton Connectivity: Unified WebSocket architecture preventing multi-connection conflicts and 406 errors.

🔬 Backtesting & Optimization

• Historical Backtesting: Simulate strategies against past data with S&P500 benchmark comparison.
• Parallel Execution: Ultra-fast multi-symbol batched backtesting using CPU parallelization.
• Grid Search Optimizer: Systematically find best parameters (Sharpe, Sortino, Calmar optimization).
• Performance Metrics: Calculates Alpha, Beta, Win Rate, and Risk-Adjusted Returns.

🛠️ Technical Stack

Core

• Language: Rust 2024 Edition 🦀
• Runtime: tokio (Asynchronous I/O, Channels, Actors)
• Database: sqlx (SQLite 0.8) for local persistence of trades and candles
• GUI: egui 0.31 + eframe (Immediate Mode GUI)

Data & Networking

• Market Data: Alpaca API v2 (WebSocket & REST), OANDA (Forex), Polygon
• WebSockets: tokio-tungstenite
• Serialization: serde, serde_json

Intelligence & Math

• Technical Indicators: ta crate + Custom Implementations (ADX, Pivots)
• Financial Math: rust_decimal (Fixed-point arithmetic for zero precision loss)

⚙️ Configuration

The application is configured primarily via environment variables. You can set these in your shell or use a .env file in the project root.

Core & Connectivity

Risk Management

...