Plug-and-play System for Scanning and Intelligent Diagnosis of Residential Network Vulnerabilities

Cybersecurity has become increasingly critical as the frequency of cyber attacks continues to rise. While critical infrastructures adopt measures to mitigate risks, the lack of protection in residential networks is often underestimated. This issue becomes more alarming as the number of reported vulnerabilities grows in parallel with the proliferation of Internet-connected devices and the expansion of the Internet of Things (IoT). In this context, this paper presents a plug-and-play vulnerability scanner built on a Raspberry Pi, designed to assist in identifying and addressing security flaws in residential environments. The proposed system integrates a Large Language Model (LLM) to generate user-friendly guidance for individuals without technical expertise. Its operation relies on three components: Nmap, used to detect devices and services within the network; the National Vulnerability Database (NVD) API, employed to retrieve Common Vulnerabilities and Exposures (CVE) associated with the detected services; and a Gemini client, which processes the retrieved CVE data to produce explanatory texts and practical remediation instructions. The system was validated in a scenario using a vulnerable target device running Metasploitable3. The results demonstrated that the application successfully identified multiple services and corresponding vulnerabilities, confirming its effectiveness in detecting potential threats. The significance of this implementation lies in its ability to present publicly available vulnerability information in a clear and accessible manner, contributing to the prevention of personal data exposure through compromised residential devices.