Article:Don Crowley
WebMCP and Consent Theatre: The Security Risks of Browser-Based AI Agents
WebMCP allows AI agents to control web applications using an active authenticated browser session. Author Don Crowley argues that current consent dialogues provide a false sense of security, as user-facing labels are decoupled from the code that actually executes. He demonstrates vulnerabilities such as tool swapping and hidden call chaining, proposing a robust three-rule handshake to secure agent actions.
From X post: WebMCP: Kieran Klaassen Demonstrates Web Integration in Open-Source App Thinkroom