Semi-automatically Augmenting Attack Trees Using an Annotated Attack Tree Library

Abstract

We present a method for assisting the semi-automatic creation of attack trees. Our method allows to explore a library of attack trees, select elements from this library that can be attached to an attack tree in construction, and determine how the attachment should be done. The process is supported by a predicate-based formal annotation of attack trees. To show the feasibility of our approach, we describe the process for automatically building a library of annotated attack trees from standard vulnerability descriptions in a publicly available online resource, using information extraction techniques. Then, we show how attack trees manually constructed from high level definitions of attack patterns can be augmented by attaching trees from this library.