A server- and browser-transparent CSRF defense for web 2.0 applications

Abstract

Cross-Site Request Forgery (CSRF) vulnerabilities constitute one of the most serious web application vulnerabilities, ranking fourth in the CWE/SANS Top 25 Most Dangerous Software Errors. By exploiting this vulnerability, an attacker can submit requests to a web application using a victim user's credentials. A successful attack can lead to compromised accounts, stolen bank funds or information leaks. This paper presents a new server-side defense against CSRF attacks. Our solution, called jCSRF, operates as a serverside proxy, and does not require any server or browser modifications. Thus, it can be deployed by a site administrator without requiring access to web application source code, or the need to understand it. Moreover, protection is achieved without requiring web-site users to make use of a specific browser or a browser plug-in. Unlike previous server-side solutions, jCSRF addresses two key aspects of Web 2.0: extensive use of client-side scripts that can create requests to URLs that do not appear in the HTML page returned to the client; and services provided by two or more collaborating web sites that need to make cross-domain requests.

WAVES: Automatic Synthesis of Client-Side Validation Code for Web Applications

“…The application is unaware that the tokens are being inserted and validated by the proxy. Proxy-based CSRF protection and WAVES do not interfere with one another as long as the proxy properly addresses AJAX requests, such as [14]. The fact that WAVES outfitted the original application to include additional JavaScript on the client and AJAX stubs on the server is irrelevant to the proxy; the developer could have added that code herself.…”