Bitget today announced a major upgrade to its Agent Hub, which gathers AI-era crypto trading infrastructure for developers and Vibe Coders. The upgrade introduces two core modules, Skills and a command-line interface (CLI), and completes a full invocation stack alongside the MCP support and REST/WebSocket APIs launched last month. Together, MCP + API + Skills + CLI form an end-to-end pathway that connects AI models, developer tooling, and real trading execution, enabling developers and AI agents to access market data, run strategies, and execute trades with significantly less friction. The newly-launched Skills mechanism lets AI agents such as Claude Code and OpenClaw automatically interpret users' trading intent and trigger real-time actions via bgc, which exposes the full API suite with standardised JSON output, enabling smooth integration for scripts, data pipelines, and system-level AI agent workflows; so agents can move from analysis to execution with minimal setup. Following the upgrade, Bitget Agent Hub now supports nine core capability modules and 58 tools. Developers can connect Bitget Agent Hub to OpenClaw in about three minutes via a simple three-step configuration. Once connected, AI agents can be granted direct access to real-time market data, spot and futures trading, account and asset management, and more, enabling intelligent trading agents that autonomously monitor markets, execute strategies, and place trades, moving toward truly AI-driven on-chain and off-chain trading. "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, CEO at Bitget. In the previous upgrade in mid-February 2026, Bitget natively supported the Model Context Protocol (MCP), enabling AI models to directly invoke Bitget trading capabilities. Through MCP, AI agents can both consume real-time market data and execute live trading actions, moving from market understanding to real participation. This is complemented by Bitget’s comprehensive, low-latency REST + WebSocket API suite spanning spot, futures, and copy trading, giving developers a reliable foundation to build automated trading systems, quantitative strategies, and AI-native trading applications. Bitget is building an AI-native trading infrastructure where Agent Hub serves as a unified entry point for AI to trade on an exchange, supporting a complete chain from natural-language instructions to live order execution, and from a single interface to an extensible ecosystem. The post Bitget upgrades Agent Hub with new tools, enabling OpenClaw trading in minutes appeared first on Invezz
