WebMCP has two APIs that allow browser agents to take action on behalf of the user.
WebMCP: A Web Standard for AI Agents
WebMCP for Browser Automation
WebMCP is a proposed web standard that lets websites declare their interactive capabilities in a format AI agents can understand and use.
Why It Matters
Traditional browser automation relies on agents scraping the DOM and clicking elements — error-prone and brittle. WebMCP eliminates this by providing:
Explicit tool discovery — agents can find and understand what actions a site supports
Structured I/O — clear JSON schemas for inputs and expected outputs, reducing hallucination
Shared state awareness — agents understand page context and can make better decisions
This is particularly important for autonomous agents and browser-based AI tools that need to interact reliably with web applications.
Current Status
Available through Chrome’s origin trial starting with version 149. Represents an active development process with community feedback shaping the standard.
Perfect for teams building agentic automation, AI browser extensions, or enterprise automation platforms.