Google Chrome has unveiled a new standard called WebMCP – Web Model Context Protocol. Developed collaboratively by Google and Microsoft engineers, and incubated through the W3C’s Web Machine Learning community group, WebMCP aims to transform how AI agents interact with websites.
WebMCP enables websites to expose structured, callable tools directly to AI agents through a new browser API: navigator.modelContext. This addresses the inefficiencies and reliability issues of traditional web-agent interactions, such as visual screen-scraping and DOM parsing.
The standard proposes two APIs – the Declarative API for standard actions defined in HTML forms, and the Imperative API for more complex interactions requiring JavaScript execution. By allowing AI agents to make structured function calls through WebMCP, the need for multiple browser interactions is reduced, leading to cost savings, improved reliability, and accelerated development for organizations embracing AI-powered applications.
Unlike fully autonomous agent paradigms, WebMCP emphasizes human-in-the-loop workflows, fostering collaborative browsing experiences where users actively participate alongside AI assistants. The standard’s design promotes cooperation between users and agents, enhancing the overall browsing experience.
WebMCP is not intended to replace existing protocols like Anthropic’s Model Context Protocol, but rather to complement them. It operates entirely client-side within the browser, offering a unique approach to web interactions that benefit from shared visual context.
Currently available in Chrome 146 Canary, WebMCP is poised to become a standard interface for AI agent interactions with the web, simplifying the current landscape of bespoke scraping strategies and automation scripts. Browser vendors and web developers alike are expected to adopt this new standard, signaling a shift in web technology.
Source: VentureBeat