OctoBot

OctoBot automates cryptocurrency trading strategies from a graphical interface, so the work is configuring and testing a bot rather than writing execution code. Written in Python and developed as free software since 2018, it runs on a home computer, a server or a cloud provider.

The latest release defaults to node mode, where your own machine acts as the backend for a new browser and mobile interface that manages wallets, automations and several strategies from one dashboard. That interface is still in beta and incomplete in places. A manual instance covers the classic workflows instead.

Strategies are assembled from built-in parts rather than written from nothing.

• Configurable strategies: grids, DCA and crypto baskets ship with the bot and are tuned from the interface.
• AI connectors: trade using an OpenAI model or any custom model served from an Ollama server.
• Signal sources: TradingView indicators, technical indicators such as RSI and MACD, and social data from Google Trends or Reddit.
• Backtesting: the built-in engine replays historical exchange data over a long period, in manual instances only.
• Exchange coverage: more than fifteen venues through the CCXT library, spot and futures depending on the exchange.

Paper trading sits between a backtest and real money, letting a strategy run live with simulated funds first. Self custody is one of the stated pillars of the new design, with keys kept on your devices and data encrypted end to end, and a Telegram bot remains available for following a running instance.