Learn the different purposes of WebMCP and MCP, and how to use them together for agentic experiences.

Chrome's platform is a leading web browser developed by Google, offering insights into web standards, browser features, and web development tools. Through documentation, developer tools, and webinars, Chrome provides insights into building modern web applications and optimizing performance. Developers can learn about web APIs, progressive web apps (PWAs), and Chrome DevTools to create fast, secure, and user-friendly web experiences.

Chrome Developers

WebMCP and MCP serve complementary roles in building agentic web experiences. MCP is a universal backend protocol connecting AI agents to external systems, data sources, and workflows across any platform. WebMCP is a proposed browser standard that exposes frontend tools to browser-based agents, giving them structured, reliable access to live website UI, DOM, session data, and cookies. Key differences: MCP is persistent and platform-agnostic; WebMCP is ephemeral and tab-bound. The recommended approach is to use MCP for core business logic and background tasks, and WebMCP for contextual in-browser interactions when a user is actively on your site.

When to use WebMCP and MCP

Thats like saying “When to use Java and Javascript”
Two totally separate things who just share the same name.

The “control how agents access your website” framing is the right lens here. MCP is great for dedicated integrations where you control both ends, but WebMCP fills the gap for surfaces where you want any agent to be able to interact with your product without requiring upfront tooling setup. The practical distinction I keep coming back to: MCP is opt-in by the developer building the integration, WebMCP is opt-in by the site owner exposing the surface. Very different deployment dynamics.

Well this does sound like it’s going to be fun to play around with. MCPs have been pretty great so far, let’s see what we can do with this.