Crypto exchange Gemini has launched a new feature called Agentic Trading, marking the first time a regulated U.S. exchange has offered direct AI agent integration for automated trading.
The platform operates through the Model Context Protocol (MCP), an open standard that provides AI agents with direct API access to execute actions on behalf of users. Gemini integrated its entire trading API with MCP, enabling AI models to access all exchange features.
The system includes modular Trading Skills — pre-built functions that AI agents call to perform specific tasks. The exchange launched with three initial modules: Get Market Data for real-time price information, Find the Spread for bid-ask analysis, and Retrieve Candles for historical price data.
Traders can connect any MCP-compatible AI model, including Anthropic's Claude and OpenAI's ChatGPT, to execute strategies ranging from basic buy and sell orders to complex multi-leg positions.
The exchange positioned the launch as part of a broader transformation in financial market interaction. "We believe we're at the beginning of a fundamental shift in how people interact with financial markets," Gemini wrote in a blog post. "Agentic trading isn't just a feature. It's a new paradigm where AI handles the execution, patterns, and discipline, while you focus on strategy and goals."
In February, Gemini announced it would slash 25% of its workforce as it streamlined operations, abandoning its businesses in the European Union, United Kingdom, and Australia to sharpen its focus on the United States. The exchange said it would boost its use of AI to become more efficient with a smaller team.
Gemini's stock is up slightly on the day, rising about 0.25% since the opening bell to recently trade at $4.40. While up about 7% in the last month, GEMI shares have fallen more than 55% since the start of the year.
Other protocols are building similar bridges between AI and crypto infrastructure. The x402 protocol, incubated by Coinbase and now under the Linux Foundation, provides AI bots access to crypto wallets and tools. Meanwhile, the Machine Payments Protocol — developed by the Stripe-backed Tempo network — enables automated machine-to-machine payments.
Why it matters
MCP gives AI agents direct API access to a regulated exchange, meaning automated strategies can be executed without manual order entry — a meaningful change in how retail and professional traders can delegate execution.
The modular Trading Skills design lets users mix and match capabilities (market data, spread analysis, historical prices), so traders can tailor automation to specific strategies rather than relying on a one-size-fits-all system.
The simultaneous emergence of x402 and the Machine Payments Protocol alongside Gemini's MCP integration indicates that multiple independent approaches to AI-to-crypto infrastructure are emerging simultaneously, and traders should evaluate which standard their preferred tools support.
