I. Introduction
Web vulnerabilities such as Cross Site Scripting (XSS) and SQL injections have been among the most targeted ones for several years [1]. Given the complexity of modern web based applications, naive blackbox fuzzing approches may not be sufficient to detect deeply nested vulnerabilities [2]. Indeed one of the problems of traditional fuzz testing is that it focuses more on data than on state transition. In the realm of blackbox testing, observing the state transition and generating inputs that traverse those states are not straightforward, which makes fuzzing less effective.