AI agents and the web – A proposal to keep developers in the loop
September 5, 2025
WebMCP
At its heart, WebMCP lets you list the actions (called “tools”) that an AI agent can perform on a page, as JavaScript functions registered via a browser API.
For example, if your web app is a media player, you could register actions such as previous, next, play, pause, or search with WebMCP, to let AI agents use these actions via the browser. This way, not only would a real person continue to be able to use your media player, by interacting with the UI as normal, but an AI agent would also be able to detect the actions that are available, and invoke them, by calling into your webapp’s code directly.
Source: Patrick – AI agents and the web – A proposal to keep developers in the loop
WebMCP is a new proposal from the Chrome and Edge teams to essentially turn web pages into MCP servers. It’s early days, but take a look at what they’re proposing and think about how it could be beneficial to what you build.
