MAITS/BANDA

The platform is written in Python and operates in a multi-threaded architecture managed by a centralized Orchestrator. It simulates multi-agent algorithmic trading (Paper Trading mode) while enforcing strict operational constraints through specialized modules: Multi-Threaded Orchestrator: Manages individual, isolated Trading Units (bots), ensuring concurrency and stability. Security Sentinel: Processes and validates all trade intents before execution to prevent out-of-bounds order sizes or unauthorized operations. Data Optimizer: Calculates real-time financial metrics, risk exposure, and tracks critical analytics such as Max Drawdown for every active bot. BandConnector: Integrates directly with Band Protocol oracles to fetch decentralized price feeds. These feeds are used as a ground-truth reference to audit local exchange data, detect price anomalies, and prevent execution based on manipulated or stale data. Logging and Auditing (Логування та аудит) The system separates operational concerns into distinct file structures to maintain clean state tracking: Bot Logs: Isolated runtime logs for individual bots are written to /home/ubuntu/kucoin_prod/bots/logs/n_bot.log (where n is the bot ID). Analytical Ledger: Trade history and portfolio metrics are committed to JSON Lines format at state/trades_history.lsonl for each bot container. System Auditor: A standalone tool (auditor.py) evaluates historical logs and real-time state to compile performance reports.