Using static analysis for Ajax intrusion detection

  • Authors:
  • Arjun Guha;Shriram Krishnamurthi;Trevor Jim

  • Affiliations:
  • Brown University, Providence, RI, USA;Brown University, Providence, RI, USA;AT&T Labs-Research, Florham Park, NJ, USA

  • Venue:
  • Proceedings of the 18th international conference on World wide web
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a static control-flow analysis for JavaScript programs running in a web browser. Our analysis tackles numerous challenges posed by modern web applications including asynchronous communication, frameworks, and dynamic code generation. We use our analysis to extract a model of expected client behavior as seen from the server, and build an intrusion-prevention proxy for the server: the proxy intercepts client requests and disables those that do not meet the expected behavior. We insert random asynchronous requests to foil mimicry attacks. Finally, we evaluate our technique against several real applications and show that it protects against an attack in a widely-used web application.