An increasing number of rich Web applications, often called AJAX applications, make use of JavaScript as a data transport mechanism. This paper describes a vulnerability we term JavaScript Hijacking, which allows an unauthorized party to read sensitive data contained in JavaScript messages.

The attack works by using a script tag to circumvent the Same Origin Policy enforced by Web browsers. Traditional Web applications are not vulnerable because they do not use JavaScript as a data transport mechanism.

We analyzed the 12 most popular AJAX frameworks, including 4 server-integrated toolkits – Direct Web Remoting (DWR), Microsoft ASP.NET AJAX (a.k.a. Atlas), XAJAX and Google Web Toolkit (GWT) -- and 8 purely client-side libraries -- Prototype, Script.aculo.us, Dojo, Moo.fx, jQuery, Yahoo! UI, Rico, and MochiKit. We determined that among them only DWR 2.0 implements mechanisms for preventing JavaScript Hijacking. The rest of the frameworks do not explicitly provide any protection and do not mention any security concerns in their documentation.

Many programmers are not using any of these frameworks, but based on our findings with the frameworks, we believe that many custom-built applications are also vulnerable. An application may be vulnerable if it:


• Uses JavaScript as a data transfer format

• Handles sensitive data

We advocate a two-pronged mitigation approach that allows applications to decline malicious requests and prevent attackers from directly executing JavaScript the applications generate.

Download the paper in PDF format here.