BlockBeats News, March 14th, developer Petr Baudis stated that with the official release of Google Chrome version 146, users can now, via a toggle, expose their current real-time browsing session to an AI Agent through the Model Context Protocol (MCP), allowing it to perform actions directly in the user's logged-in browsing environment.
Baudis also revealed that he had previously envisioned enabling Claude to browse and perform tasks on behalf of the user in a logged-in state through the command amp mcp add chrome-devtools -- npx chrome-devtools-mcp@latest --autoConnect. Now, with Chrome's MCP capability in place, this scenario has gradually become a reality.
