AI Browser Guard Exposes When Bots Are Driving Your Browser

As automated browsing by AI agents becomes more common, a new Chrome extension called “AI Browser Guard” is gaining traction among developers and security-conscious users. The extension is designed to detect when tools like Playwright, Puppeteer, Selenium, or newer agent frameworks from Anthropic and OpenAI take control of your browser, and then surface that activity in a clear status panel.

AI Browser Guard combines several signals — WebDriver flags, Chrome DevTools Protocol markers, and behavioral patterns in mouse and keyboard input — to estimate whether a human or an automated agent is driving the session. It shows a confidence score and the detection method, and it logs observed actions in a session timeline so you can later review what scripts did on which sites.

An optional feature also reads a site’s /.well-known/ai-safety.txt file, if present, to display how that domain self-describes its safety and AI usage practices. The developer notes that a browser extension can’t fully enforce policy against frameworks that control the debugging interface, but argues that visibility is the first step toward safer agentic automation. With enterprises reassessing the risks of embedded AI in browsers, the tool is being discussed as a lightweight “AI firewall” for everyday browsing.