Walmart unveils Agent Gateway, an MCP server for AI agents to shop its site, and Agent Profiles to authorize agents and track their purchases separately

Shopping for items using an AI agent now has a sanctioned, controllable pathway on Walmart.com.

Walmart has launched "Agent Gateway," a new system designed to allow AI agents to shop on its website and at Sam's Club. This move contrasts with Amazon's recent decision to block AI agents. The Agent Gateway utilizes the Model Context Protocol (MCP) to enable approved AI agents to search Walmart's inventory and complete purchases on behalf of users. To enhance customer control and security, Walmart is also introducing "Agent Profiles." These profiles allow customers to authorize specific AI agents, track their purchases separately from personal orders, and set spending limits or require approval for each transaction. These features aim to prepare Walmart for the mainstream adoption of AI agents in commerce, offering a secure and controlled environment for agentic shopping. Agents must comply with specific terms of service, and non-compliant agents will be blocked from the site.

Walmart's proactive embrace of agentic commerce, through its Agent Gateway and Agent Profiles, signals a strategic pivot to capitalize on the evolving landscape of AI-driven consumer behavior. By offering a regulated environment, Walmart differentiates itself from competitors like Amazon, which are opting for exclusion. This move could set a precedent for how other retailers integrate with AI agents, potentially normalizing agentic shopping and encouraging broader adoption across the e-commerce sector.

The introduction of MCP-based communication and granular controls within Agent Profiles addresses key concerns around data privacy, security, and user autonomy. This approach not only builds trust for consumers experimenting with AI agents for purchases but also provides retailers with necessary oversight. The success of Walmart's strategy may influence the development of industry standards for agent-to-retailer interactions and could accelerate the development of more sophisticated, trustworthy AI shopping assistants.