Crypto exchange Gemini just introduced Agentic Trading, a feature that connects AI models directly to user trading accounts. The tool allows AI agents built on Claude and ChatGPT to monitor markets, execute trades, and manage risk autonomously. Gemini calls it “the first agentic trading tool to be available directly through a regulated US-based exchange.” The launch signals a major shift in how traders interact with centralized exchanges. Instead of manually placing orders, users can now delegate trading decisions to AI.
Agentic trading is now live on Gemini 🤖
— Gemini (@Gemini) April 18, 2026
Connect your AI model of choice directly to the exchange, train it with plain language, and run your own trading systems directly on the exchange
Built to get you trading in minutes pic.twitter.com/cU0aISwveu
How Gemini Agentic Trading Works
Agentic Trading operates on the Model Context Protocol (MCP), an open standard originally developed by Anthropic. MCP provides a universal framework for connecting AI models to external tools, APIs, and data sources. Gemini has integrated its full trading API with MCP, creating a bridge between AI agents and exchange functionality. This means any MCP-compatible AI model can access Gemini’s trading infrastructure natively. Users connect their preferred AI model to their Gemini account, define trading strategies, and let the agent operate within those parameters.
The system supports both Claude, built by Anthropic, and ChatGPT, built by OpenAI. These models can interpret market conditions, analyze historical data, and execute trades based on predefined strategies. Importantly, human oversight remains part of the design. Users set the rules, and AI agents operate within those boundaries.
Calling all robots 🤖. Check out @Gemini's new MCP API, designed specifically for AI agents to trade predictions, crypto, and more. https://t.co/BQcLhgpOse
— Tyler Winklevoss (@tyler) April 20, 2026
Trading Skills: Pre-Built Modules for AI Agents
Gemini also introduced “Trading Skills,” a set of pre-built modular functions that AI agents can call during operation. These include tools like Find the Spread, which queries the bid-ask spread for any trading pair on the exchange. Another module, Retrieve Candles, gives agents access to historical candlestick data for pattern recognition and backtesting.
These Trading Skills lower the barrier for users who want AI-driven trading but lack the technical background to build custom integrations. Rather than writing API calls from scratch, users can leverage these ready-made functions. The modular approach also means Gemini can expand the toolkit over time without overhauling the core system. For developers, Trading Skills offer building blocks that accelerate the creation of more sophisticated trading agents.
Why MCP Matters for Crypto Trading
MCP has quickly become the connective tissue between AI models and external services. Anthropic released the protocol in late 2024 as an open standard. Since then, OpenAI, Google DeepMind, Microsoft, and others have adopted it. Anthropic later donated MCP to the Agentic AI Foundation, a directed fund under the Linux Foundation co-founded by Anthropic, Block, and OpenAI.
For crypto exchanges, MCP solves a real integration problem. Previously, connecting AI models to trading infrastructure required custom API wrappers and proprietary setups. MCP standardizes this process, making it model-agnostic. Any AI model that supports MCP can plug into Gemini’s trading system. This interoperability matters because it prevents vendor lock-in and encourages broader adoption.
Competitive Landscape: Coinbase Moves in a Similar Direction
Gemini is not the only exchange embracing AI agent infrastructure. Coinbase recently launched its own Payments MCP tool, which enables AI agents to access blockchain wallets and conduct crypto transactions. The Coinbase tool supports Claude and Google’s Gemini AI model, and it focuses on stablecoin payments and onramps rather than active trading.
Coinbase has also partnered with Cloudflare to launch the x402 Foundation, which aims to standardize AI payments. The thesis from both exchanges is the same: stablecoins and crypto rails move at the speed of code, making them natural infrastructure for autonomous AI agents. However, Gemini’s approach targets active trading execution, while Coinbase focuses more on payments and wallet access.
What This Means for Traders and the Industry
Gemini Agentic Trading represents one of the first real-world implementations of MCP in regulated financial services. The feature turns Gemini from a traditional exchange into a platform that AI agents can operate on natively. For retail traders, this opens the door to algorithmic strategies that previously required significant technical expertise. For institutional players, it creates a standardized way to deploy AI trading agents on a compliant, regulated platform.
The broader implication extends beyond Gemini. As MCP adoption grows across exchanges, AI agents will increasingly interact with multiple platforms through a single protocol. This could reshape how liquidity flows across centralized exchanges. It also raises important questions about risk management, regulatory oversight, and the role of human decision-making in automated trading. For now, Gemini has positioned itself at the front of this shift by being the first regulated U.S. exchange to offer the capability directly.
*Disclaimer: News content provided by Genfinity is intended solely for informational purposes. While we strive to deliver accurate and up-to-date information, we do not offer financial or legal advice of any kind. Readers are encouraged to conduct their own research and consult with qualified professionals before making any financial or legal decisions. Genfinity disclaims any responsibility for actions taken based on the information presented in our articles. Our commitment is to share knowledge, foster discussion, and contribute to a better understanding of the topics covered in our articles. We advise our readers to exercise caution and diligence when seeking information or making decisions based on the content we provide.
