arXiv subjects

Angelo Furfaro

Publications and source records attributed to Angelo Furfaro.

9 recordsLinked to original sources

arXiv2025

ICSLure: A Very High Interaction Honeynet for PLC-based Industrial Control Systems

The security of Industrial Control Systems (ICSs) is critical to ensuring the safety of industrial processes and personnel. The rapid adoption of Industrial Internet of Things (IIoT) technologies has expanded system functionality but also increased the attack surface, exposing ICSs to a growing range of cyber threats. Honeypots provide a means to detect and analyze such threats by emulating target systems and capturing attacker behavior. However, traditional ICS honeypots, often limited to software-based simulations of a single Programmable Logic Controller (PLC), lack the realism required to engage sophisticated adversaries. In this work, we introduce a modular honeynet framework named ICSLure. The framework has been designed to emulate realistic ICS environments. Our approach integrates physical PLCs interacting with live data sources via industrial protocols such as Modbus and Profinet RTU, along with virtualized network components including routers, switches, and Remote Terminal Units (RTUs). The system incorporates comprehensive monitoring capabilities to collect detailed logs of attacker interactions. We demonstrate that our framework enables coherent and high-fidelity emulation of real-world industrial plants. This high-interaction environment significantly enhances the quality of threat data collected and supports advanced analysis of ICS-specific attack strategies, contributing to more effective detection and mitigation techniques.

cs.CR↗

arXiv2025

ARPaCCino: An Agentic-RAG for Policy as Code Compliance

Policy as Code (PaC) is a paradigm that encodes security and compliance policies into machine-readable formats, enabling automated enforcement in Infrastructure as Code (IaC) environments. However, its adoption is hindered by the complexity of policy languages and the risk of misconfigurations. In this work, we present ARPaCCino, an agentic system that combines Large Language Models (LLMs), Retrieval-Augmented-Generation (RAG), and tool-based validation to automate the generation and verification of PaC rules. Given natural language descriptions of the desired policies, ARPaCCino generates formal Rego rules, assesses IaC compliance, and iteratively refines the IaC configurations to ensure conformance. Thanks to its modular agentic architecture and integration with external tools and knowledge bases, ARPaCCino supports policy validation across a wide range of technologies, including niche or emerging IaC frameworks. Experimental evaluation involving a Terraform-based case study demonstrates ARPaCCino's effectiveness in generating syntactically and semantically correct policies, identifying non-compliant infrastructures, and applying corrective modifications, even when using smaller, open-weight LLMs. Our results highlight the potential of agentic RAG architectures to enhance the automation, reliability, and accessibility of PaC workflows.

cs.SE↗

arXiv2025

The Dark Side of LLMs: Agent-based Attack Vectors for System-level Compromise

The rapid adoption of Large Language Model (LLM) agents and multi-agent systems enables remarkable capabilities in natural language processing and generation. However, these systems introduce security vulnerabilities that extend beyond traditional content generation to system-level compromises. This paper presents a comprehensive evaluation of the LLMs security used as reasoning engines within autonomous agents, highlighting how they can be exploited as attack vectors capable of achieving computer takeovers. We focus on how different attack surfaces and trust boundaries can be leveraged to orchestrate such takeovers. We demonstrate that adversaries can effectively coerce popular LLMs into autonomously installing and executing malware on victim machines. Our evaluation of 18 state-of-the-art LLMs reveals that 94.4% of models succumb to Direct Prompt Injection, and 83.3% are vulnerable to the more stealthy and evasive RAG Backdoor Attack. Notably, we tested trust boundaries within multi-agent systems, where LLM agents interact and influence each other, and we revealed that LLMs which successfully resist direct injection or RAG backdoor attacks will execute identical payloads when requested by peer agents. We found that 100.0% of tested LLMs can be compromised through Inter-Agent Trust Exploitation attacks, and that every model exhibits context-dependent security behaviors that create exploitable blind spots.

cs.CR↗

arXiv2025

CyberRAG: An Agentic RAG cyber attack classification and reporting tool

Intrusion Detection and Prevention Systems (IDS/IPS) in large enterprises can generate hundreds of thousands of alerts per hour, overwhelming analysts with logs requiring rapidly evolving expertise. Conventional machine-learning detectors reduce alert volume but still yield many false positives, while standard Retrieval-Augmented Generation (RAG) pipelines often retrieve irrelevant context and fail to justify predictions. We present CyberRAG, a modular agent-based RAG framework that delivers real-time classification, explanation, and structured reporting for cyber-attacks. A central LLM agent orchestrates: (i) fine-tuned classifiers specialized by attack family; (ii) tool adapters for enrichment and alerting; and (iii) an iterative retrieval-and-reason loop that queries a domain-specific knowledge base until evidence is relevant and self-consistent. Unlike traditional RAG, CyberRAG adopts an agentic design that enables dynamic control flow and adaptive reasoning. This architecture autonomously refines threat labels and natural-language justifications, reducing false positives and enhancing interpretability. It is also extensible: new attack types can be supported by adding classifiers without retraining the core agent. CyberRAG was evaluated on SQL Injection, XSS, and SSTI, achieving over 94\% accuracy per class and a final classification accuracy of 94.92\% through semantic orchestration. Generated explanations reached 0.94 in BERTScore and 4.9/5 in GPT-4-based expert evaluation, with robustness preserved against adversarial and unseen payloads. These results show that agentic, specialist-oriented RAG can combine high detection accuracy with trustworthy, SOC-ready prose, offering a flexible path toward partially automated cyber-defense workflows.

cs.CR↗

arXiv2025

ARCeR: an Agentic RAG for the Automated Definition of Cyber Ranges

The growing and evolving landscape of cybersecurity threats necessitates the development of supporting tools and platforms that allow for the creation of realistic IT environments operating within virtual, controlled settings as Cyber Ranges (CRs). CRs can be exploited for analyzing vulnerabilities and experimenting with the effectiveness of devised countermeasures, as well as serving as training environments for building cyber security skills and abilities for IT operators. This paper proposes ARCeR as an innovative solution for the automatic generation and deployment of CRs, starting from user-provided descriptions in a natural language. ARCeR relies on the Agentic RAG paradigm, which allows it to fully exploit state-of-art AI technologies. Experimental results show that ARCeR is able to successfully process prompts even in cases that LLMs or basic RAG systems are not able to cope with. Furthermore, ARCeR is able to target any CR framework provided that specific knowledge is made available to it.

cs.CR↗

arXiv2018

CalcuList: a Functional Language Extended with Imperative Features

CalcuList (Calculator with List manipulation), is an educational language for teaching functional programming extended with some imperative and side-effect features, which are enabled under explicit request by the programmer. In addition to strings and lists, the language natively supports json objects. The language adopts a Python-like syntax and enables interactive computation sessions with the user through a REPL (Read-Evaluate-Print-Loop) shell. The object code produced by a compilation is a program that will be eventually executed by the CalcuList Virtual Machine (CLVM).

cs.PL↗

arXiv2015

Honeypot-powered Malware Reverse Engineering

Honeypots, i.e. networked computer systems specially designed and crafted to mimic the normal operations of other systems while capturing and storing information about the interactions with the world outside, are a crucial technology into the study of cyber threats and attacks that propagate and occur through networks. Among them, high interaction honeypots are considered the most efficient because the attacker (whether automated or not) perceives realistic interactions with the target machine. In the case of automated attacks, propagated by malwares, currently available honeypots alone are not specialized enough to allow the analysis of their behaviors and effects on the target system. The research presented in this paper shows how high interaction honeypots can be enhanced by powering them with specific features that improve the reverse engineering activities needed to effectively analyze captured malicious entities.

cs.CR↗

arXiv2014

Exploiting n-gram location for intrusion detection

Signature-based and protocol-based intrusion detection systems (IDS) are employed as means to reveal content-based network attacks. Such systems have proven to be effective in identifying known intrusion attempts and exploits but they fail to recognize new types of attacks or carefully crafted variants of well known ones. This paper presents the design and the development of an anomaly-based IDS technique which is able to detect content-based attacks carried out over application level protocols, like HTTP and FTP. In order to identify anomalous packets, the payload is split up in chunks of equal length and the n-gram technique is used to learn which byte sequences usually appear in each chunk. The devised technique builds a different model for each pair and uses them to classify the incoming traffic. Models are build by means of a semi-supervised approach. Experimental results witness that the technique achieves an excellent accuracy with a very low false positive rate.

cs.CR↗

arXiv2014

A multi-protocol framework for the development of collaborative virtual environments

Collaborative virtual environments (CVEs) are used for collaboration and interaction of possibly many participants that may be spread over large distances. Both commercial and freely available CVEs exist today. Currently, CVEs are used already in a variety of different fields: gaming, business, education, social communication, and cooperative development. In this paper, a general framework is proposed for the development of a cooperative environment which is able to exploit a multi protocol network infrastructure. The framework offers support to concerns such as communication security and inter-protocol interoperability and let software engineers to focus on the specific business of the CVE under development. To show the framework effectiveness we consider, as a case of study, the design of a reusable software layer for the development of distributed card games built on top of it. This layer is, in turn, used for the implementation of a specific card game.

cs.SE↗