Bitget, the world's largest Universal Exchange (UEX), has announced an upgrade to its Agent Hub platform, adding new modules designed to support AI-driven trading and automated strategy execution.
The update introduces two new components, Skills and a command-line interface (CLI), that expand the infrastructure available to developers building trading applications powered by artificial intelligence. Together with previously released integrations, the update forms a complete toolset that connects AI models, developer tools, and the exchange’s trading systems.
The system now combines several layers of access, including the Model Context Protocol (MCP), REST and WebSocket APIs, Skills modules, and the CLI interface. According to the company, the goal is to simplify connecting AI models to real trading environments.
Skills modules and AI strategy
The new Skills mechanism allows AI agents to interpret trading instructions and trigger actions directly through the exchange’s infrastructure. Agents integrated with development tools such as Claude Code or OpenClaw can process user prompts, convert them into trading logic, and execute tasks through Bitget’s trading system.
Through a utility called bgc, the platform exposes exchange functions through standardized JSON outputs. This structure is intended to simplify integration with scripts, data pipelines, and automated workflows used by AI-driven systems.
Once connected, AI agents can access real-time market data, evaluate trading strategies, and submit orders with relatively limited setup compared with traditional API-based integrations.
Following the upgrade, Agent Hub now includes nine primary capability modules and 58 developer tools. Bitget states that developers can connect the platform to AI frameworks such as OpenClaw through a short configuration process.
Once integration is completed, AI agents can access functions such as spot and derivatives trading, account management, asset monitoring, and portfolio data. These tools allow automated systems to monitor market conditions, evaluate strategies, and place trades through the exchange.
The company says the framework is designed to support both on-chain and off-chain trading workflows managed by autonomous agents.
Building AI-native trading environment
The latest upgrade follows an earlier update released in February 2026, when Bitget added native support for the Model Context Protocol. MCP allows AI systems to directly invoke trading functions while also retrieving market information through standardized calls.
Combined with the exchange’s REST and WebSocket APIs, the protocol allows developers to build automated trading systems capable of gathering real-time market data and executing trades through the same environment.
According to Bitget CEO Gracy Chen, the company views AI-driven trading tools as a key part of its long-term development strategy.
"As one of the core strategic focuses this year, AI is the tool to translate intent into execution with speed and clarity. With AI, our mission is to enable 125 million users and beyond trade like Wall Street pros. With this, we provide the broadest functional coverage and security amongst major exchanges. While AI trading still requires more education and trust, the inflection point for it is approaching," said Gracy Chen.
The exchange describes Agent Hub as a unified entry point for AI trading applications, designed to support the full process from natural-language instructions to strategy execution and order placement. The platform is intended to serve both developers building trading tools and AI agents operating automated trading systems on the exchange.
Dan Burgin
Vladislav Sopov
