MetaQuotes released MetaTrader 5 Build 6180 on Thursday, September 3, 2026, extending the set of tools available to its AI Assistant.
External AI agents connected via Model Context Protocol can now retrieve reports and logs from the Strategy Tester and analyse them. They can also initiate optimisation runs directly, making it easier for traders to test and improve trading robots.
The update continues MetaQuotes' push to turn MT5 into an agentic AI platform. The company introduced native MCP support in MetaEditor and MetaTrader 5 with Build 6060, released on July 23, 2026.
AI Agents Get More Powers
Previously, the AI Assistant could analyse markets, explain trading activity, and help with Expert Advisor testing and debugging. The new build adds retrieving testing and optimisation results, configuring optimisation criteria in the Strategy Tester, checking current tester settings, and adding indicators to charts.
For example, a trader can ask the Assistant to add several moving averages with specific settings, or to adjust their parameters for better results.
The added functionality covers research, analysis, debugging, and algorithm optimisation inside the Strategy Tester. Judgment-based functionality — trading decisions on a live account — remains unavailable at this stage.
- MetaQuotes Hits 1 Trillion Tokens Since July MCP Release
- Are Third-Party Bridge Providers Being Priced Out By MetaQuotes?
- MetaQuotes Launches MetaTrader.com as Standalone Financial Data Portal
Where Competitors Go Further
MetaQuotes is connecting AI agents to its trading infrastructure through an open protocol rather than a proprietary integration. Some competitors have taken a less conservative approach, letting agents make decisions and execute trades directly.
Spotware Systems, the company behind cTrader, launched cTrader AI Agent Connect in May, built around two MCP servers and a skills library that gives agents direct access to trading operations.
The company describes it as the first built-in AI agent solution in FX/CFD trading. eToro takes automation further still: its Agent Portfolios let AI agents trade autonomously within a dedicated sub-account, inside limits the user sets.
MetaQuotes reported in August that the AI Assistant had processed more than 1 trillion tokens in its first three weeks after the July rollout of MCP support — the last usage figure the company has disclosed to date.
