MetaMask Unveils Agent Wallet for Autonomous DeFi Trading
MetaMask has introduced Agent Wallet, a self-custodial cryptocurrency wallet engineered to empower AI agents to autonomously engage in trading and interact with decentralized finance (DeFi) protocols, all while operating under user-defined security parameters.
The launch arrives as the cryptocurrency development community increasingly focuses on creating AI agents capable of managing portfolios, executing trades, and directly interacting with decentralized applications. The product is currently accessible to approximately 200 users through an Early Access Program, with a broader release anticipated later this summer.
“It’s genuinely day one for agents, but the infrastructure decision can’t wait because agents are already touching real money, and most of them are doing it the wrong way,” Zhen Yu Tong, MetaMask Senior Director of Product, told Decrypt.
Tong explained that many existing market solutions grant AI agents direct access to private keys, posing a significant risk of unintended transactions or fund loss due to errors rather than malicious hacks.
“If the first generation of trading agents normalizes giving away your keys, we’ll be rebuilding the custodial mistakes crypto spent a decade escaping,” he stated.
MetaMask’s Agent Wallet routes transactions through the company’s existing security infrastructure. This includes transaction simulation, detection of scams and malicious contracts, threat scanning powered by Blockaid, Clear Signing, and Servo MEV protection.
Rather than assuming AI models can be entirely shielded from manipulation, MetaMask emphasized that the wallet’s design prioritizes controls to mitigate the consequences of agent errors.
“The honest premise first: You cannot guarantee an LLM won’t be tricked,” Tong said. “Prompt injection is an open research problem, not a bug you patch once.”
Prompt injection attacks occur when malicious instructions are used to compromise an AI system, leading it to perform unintended actions. In the context of cryptocurrency, this could involve tricking an agent into approving transactions, moving funds, or interacting with a compromised smart contract.
To counter these risks, Agent Wallet’s default Guard Mode allows users to establish spending limits, approved protocols, and other operational parameters. Transactions that deviate from these rules or are flagged as suspicious necessitate two-factor authentication for completion.
A less restrictive Beast Mode offers agents greater autonomy, though it still requires approval for transactions identified as malicious.
“Beast Mode is for users who want genuinely hands-off operation—the agent acts without a pop-up on every transaction,” Tong explained. “What Beast Mode does not do is switch off the safety net. If our threat detection flags a transaction as malicious, 2FA still fires no matter what mode you’re in. That’s non-negotiable.”
Tong further clarified that operating without explicit approval does not equate to operating without limitations. Beast Mode still adheres to user-defined guardrails, including spending caps, approved assets and protocols, and time-based restrictions. This enables agents to autonomously rebalance portfolios, interact with verified contracts, and settle payments without requiring individual transaction approvals.
“Think of it like banks or exchanges, where you need to add recipients to an allowlist before you can send to them. That’s Guard Mode—the user pre-approves who the agent can interact with, and anything outside that list triggers 2FA,” he said. “Beast Mode flips it: addresses are scanned in real time, and 2FA fires if any are flagged as bad—but the user doesn’t have to add anyone to an allowlist upfront.”
The wallet supports Ethereum Virtual Machine-compatible chains and Hyperliquid, along with agent frameworks such as OpenAI Codex, Anthropic’s Claude Code, Cursor, OpenClaw, and Hermes Agent. It utilizes Cubist’s trusted execution environment technology to ensure private keys remain within a hardware-isolated enclave during the signing process, thereby preventing MetaMask and Consensys from accessing user key material.
MetaMask’s Agent Wallet launch follows similar initiatives from other crypto companies developing infrastructure for AI agents.
In February, Coinbase introduced Agentic Wallets, a self-custodial wallet designed for AI agents to manage crypto assets and send payments, with private keys isolated within trusted execution environments. In March, MoonPay expanded its agent strategy by integrating Ledger hardware wallets for AI transactions requiring human approval.
The crypto payments firm later launched the Open Wallet Standard, an open-source framework supported by entities including PayPal, the Ethereum Foundation, Solana Foundation, Ripple, and Base, aiming to standardize AI agent wallet and fund management across blockchains.
Last week, MoonPay released a desktop application for Claude Code and OpenAI Codex, enabling users to connect AI assistants to wallets, token swaps, prediction markets, and other blockchain tools through a graphical interface.
Leave a Reply