honvu93/chronostrade

ChronosTrade is a self-hosted trading platform built for people who write algorithmic trading strategies and want to run them against real markets. It connects to MetaTrader 5, a widely used brokerage terminal, through a Python bridge that fetches historical price data and sends live trade orders. The rest of the system is built in TypeScript and runs locally on your own machine, meaning your strategies, API keys, and trade data never leave your control. The platform covers the full workflow from data to execution. Price candles from MetaTrader 5 get stored in a time-series database called TimescaleDB, which is designed to handle large amounts of financial data efficiently. From there, the system computes technical indicators on the stored data. Technical indicators are mathematical formulas traders use to spot patterns in price movement, such as RSI (which measures how overbought or oversold an asset is) or EMA (which smooths out price trends over time). You configure these directly in a web dashboard without editing any code. Backtesting is a core feature. That means you can replay a strategy against historical price data to see how it would have performed before risking real money. ChronosTrade runs backtests in parallel using a queue system, so sweeping through many combinations of indicator settings to find the best one does not require waiting hours for results to come back one at a time. For live trading, the platform includes a set of risk controls called trade guards. These can pause trading after a run of consecutive losses, cap total daily losses, or filter entries based on whether your account equity is trending in the right direction. Exit strategies are also configurable: you can set fixed profit targets, move the stop loss to break-even once a trade is profitable, close part of a position early, or trail the stop based on recent price swings. A Next.js dashboard ties everything together. You can watch trades and indicator signals arrive in real time, inspect indicator values bar by bar on professional charts, and adjust parameters on the fly. Telegram notifications are also supported for signals and events.