arXiv subjects

Rui Claro

Publications and source records attributed to Rui Claro.

3 recordsLinked to original sources

arXiv2026

RedShell: A Generative AI-Based Approach to Ethical Hacking

The application of Machine Learning techniques in code generation is now a common practice for most developers. Tools such as ChatGPT from OpenAI leverage the natural language processing capabilities of Large Language Models to generate machine code from natural language descriptions. In the cybersecurity field, red teams can also take advantage of generative models to build malicious code generators, providing more automation to Pentest audits. However, the application of Large Language Models in malicious code generation remains challenging due to the lack of data to train and evaluate offensive code generators. In this work, we propose RedShell, a tool that allows ethical hackers to generate malicious PowerShell code. We also introduce a ground truth dataset, combining publicly available code samples to fine-tune models in malicious PowerShell generation. Our experiments demonstrate the strong capabilities of RedShell in generating syntactically valid PowerShell, with fewer than 10% of the generated samples resulting in parse errors. Furthermore, our specialized model was able to produce samples that were semantically consistent with reference snippets, achieving a competitive performance on standard output similarity metrics such as Edit Distance and METEOR, with their mean similarity scores exceeding 50% and 40%, respectively. This work sheds light on the state-of-the-art research in the field of Generative AI applied to Pentesting, and also serves as a steppingstone for future advancements, highlighting the potential benefits these models hold within such controlled environments.

cs.CR↗

arXiv2026

Towards Automated Pentesting with Large Language Models

Large Language Models (LLMs) are redefining offensive cybersecurity by allowing the generation of harmful machine code with minimal human intervention. While attackers take advantage of dark LLMs such as XXXGPT and WolfGPT to produce malicious code, ethical hackers can follow similar approaches to automate traditional pentesting workflows. In this work, we present RedShell, a privacy-preserving, hardware-efficient framework that leverages fine-tuned LLMs to assist pentesters in generating offensive PowerShell code targeting Microsoft Windows vulnerabilities. RedShell was trained on a malicious PowerShell dataset from the literature, which we further enhanced with manually curated code samples. Experiments show that our framework achieves over 90% syntactic validity in generated samples and strong semantic alignment with reference pentesting snippets, outperforming state-of-the-art counterparts in distance metrics such as edit distance (above 50% average code similarity). Additionally, functional experiments emphasize the execution reliability of the snippets produced by RedShell in a testing scenario that mirrors real-world settings. This work sheds light on the state-of-the-art research in the field of Generative AI applied to malicious code generation and automated testing, acknowledging the potential benefits that LLMs hold within controlled environments such as pentesting.

cs.CR↗

arXiv2022

Lisbon Hotspots: Wi-Fi access point dataset for time-bound location proofs

Wi-Fi hotspots are a valuable resource for people on the go, especially tourists, as they provide a means to connect personal devices to the Internet. This extra connectivity can be helpful in many situations, e.g., to enable map and chat applications to operate outdoors when cellular connectivity is unavailable or is expensive. Retail stores and many public services have recognized that hotspots have potential to attract and retain customers, so many of them offer free and open Wi-Fi. In busy cities, with many locals and visitors, the number of hotspots is very significant. Some of these hotspots are available for long periods of time, while others are short-lived. When we have many users with devices collecting hotspot observations, they can be used to detect the location -- using the long-lived hotspots -- and to prove the time when the location was visited -- using the short-lived hotspots observed by others users at the location. In this article, we present a dataset of collected Wi-Fi data from the most important tourist locations in the city of Lisbon, Portugal, over a period of months, that was used to show the feasibility of using hotspot data for location detection and proof. The obtained data and algorithms were assessed for a specific use case: smart tourism. We also present the data model used to store the observations and the algorithms developed to detect and prove location of a user device at a specific time. The Lisbon Hotspots dataset, LXspots, is made publicly available to the scientific community so that other researchers can also make use of it to develop new and innovative mobile and Internet of Things applications.

cs.CY↗