Coinbase’s Ethereum layer-2 network, Base, has launched a tool enabling AI agents like Claude and ChatGPT to perform blockchain operations. The Base MCP allows users to ask agents to transfer funds, swap tokens, and use apps through a chat interface. Each proposed transaction requires manual confirmation in the wallet, and private keys remain secure. The tool also expands adoption of Coinbase’s x402 agentic payment protocol, though current transaction volume remains modest.
Base, the Ethereum layer-2 blockchain from crypto exchange Coinbase, launched a new tool to connect Base accounts to artificial intelligence agents. The Base MCP (Model Context Protocol) allows users to ask AI agents such as Anthropic’s Claude or OpenAI’s ChatGPT to transfer funds, swap tokens, check balances, and use supported apps. Users can manage crypto directly from an AI model’s chat interface and interact with protocols like Morpho, Moonwell, and Uniswap.
The AI agent proposes an action in the chat, then the Base wallet opens for user confirmation. The agent does not have access to private keys, and every action must be confirmed. Lincoln Murr, head of AI Product for Coinbase, stated that “unlike siloed agentic wallets that only live in a terminal, your Base Account travels with you — trades, history, and portfolio sync whether you’re in-agent or in the Base App.” Base MCP will also expand adoption of the Coinbase x402 protocol, an agentic AI payment standard launched in May 2025.
Murr described MCP as a “nice wrapper” on top of APIs. Together with x402, it enables a new micro-transaction economy where agents can make tiny payments in crypto. However, that economy is currently in its early stages, with x402 processing just $1.1 million in volume over the past 30 days, according to x402scan. The use of AI agents for crypto payments has its detractors, with a recent research paper saying AI agents should be treated as an untrusted system component.
The researchers said AI agents should clearly distinguish between instructions and untrusted data to avoid attackers duping the agent. Earlier this week, the developer platform Socket discovered malware targeting crypto developers by injecting hidden instructions to hijack AI coding assistants. Backers argue AI models will have issues accessing the banking system and will need to use digital assets to transact, touting agentic payments as the next major crypto use case.
