Traders are starting to talk to their platforms instead of just clicking on them. cTrader has rolled out an official AI agent connection that lets natural-language prompts review markets, check accounts, and even trigger trades. The connection runs on the Model Context Protocol, or MCP, a system built to link AI apps with trading functions. It works with several AI clients, including Claude Code, Codex, Cursor, Windsurf, Gemini CLI, and GitHub Copilot. cTrader calls this AI-assisted trading, where a simple prompt can lead to real platform actions like placing an order or managing a position.
cTrader now lets traders talk instead of click, using MCP to turn natural-language prompts into real trading actions.
The setup relies on two MCP servers. One is remote, built for cTrader Web, and it handles trading, account data, and market data. The other is local, made for Windows or Mac, and it does much more. It covers trading, market data, account details, chart control, indicators, plugins, UI layout, workspaces, and price alerts. The local version gives the AI agent direct access to live trading, market data, and even screen control. Getting started means connecting a supported AI client and switching on MCP in cTrader's settings, using a token or configuration snippet.
cTrader recommends starting slow. The official path begins with a safe prompt, usually on a demo account (Official Site 🔗), often in read-only mode before real trading gets switched on. Once enabled, the AI agent can place, modify, and close actual orders. It can handle market, limit, stop, and stop-limit orders, along with managing pending orders and closing positions. The local server can also be set to ask for confirmation before every trade, adding a safety check. Since cTrader operates as a multi-dealer platform aggregating liquidity from several institutions, AI agents executing trades through it may interact with pricing drawn from multiple sources rather than a single dealing desk.
This matters because it shifts AI from simple chart-reading and news summaries toward direct execution, cutting the gap between analysis and action. But cTrader is clear that supervision still matters. Users must manually turn on trading permissions, and the platform advises checking AI outputs before acting on them.
For now, this is a platform-level feature rather than a market-moving event. Still, traders watching currencies, commodities, or African markets may find AI agents useful for faster account checks and structured order execution, provided human oversight stays part of the process.
