Artifact for FM'24 paper: Fast Attack Graph Defense Localization via Bisimulation

Supporting Materials : The supporting materials include the following folders: Datasets: a folder containing input files storing security findings of datasets I-IV. The findings are encoded as input.P files, and hold predicate assignments. The analytic attack graphs described in the paper are generated by running attack graph simulator tool. The tool take input.P file and a file containing security rules encoded in datalog, and outputs and attack graphs. - SET I: input files of security findings of two large manufacturing facilities in the automotive and retail industry. - SET II: input files of security findings of three of the IT attack graphs - SET III: input files of security findings of the RE20 datasets - SET IV: input files of security findings of simulated attack graphs (small, medium, large, xl) Results: Results.csv: raw results of experiments presented in the evaluation section. Figures: figures that appear in the paper. Scripts (Folder): • Readme.md: instructions on how to run the Analysis.py • Analysis.py: script used to create the tables and figure in the evaluation section from Results.csv • Readme.md: instructions on how to run the Analysis.py • requirements.txt: list of packages required to run Analysis.py To recreate the paper tables and figures, please follow the instructions in the readme.txt Important notes, due to IP restrictions we are unable to share the following artifacts: • Security rules file containing security rules in Datalog, which were used to construct that attack graphs. • Algorithm for constructing and folding the AAGs. • Analytic Attack graphs (AAG) (as these reveal the security rules)