Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{187003,
        author = {Amol Dhankar and Sakshi Jalit and Jayant Kondekar and Chandan Singh and Vaishnavi Shukla and Ujjwal Jamaiwar and Mayur Ingole},
        title = {PhishGuard Extension: Real-Time Phishing Detector with Smart Language Analysis},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {3527-3531},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187003},
        abstract = {The project uses the DistilBERT model, which aims to enhance semantic and contextual understanding in phishing content analysis.
Phishing attacks have been among the most prevalent and damaging forms of cybercrime, exploiting user trust to steal users’ sensitive information such as passwords and financial details and personal data. Traditional blacklist and rule-based detection techniques often fail to identify newly generated or sophisticated phishing websites. As a result, this paper proposes the Browser-Extension Phishing Detection directly connected to Chrome, using Machine Learning (ML) and Natural Language Processing (NLP) to endow the browser with real-time phishing detection intelligence. The system will embed an optimized DistilBERT classifier hosted on the FastAPI backend, seamlessly communicating with the Chrome Extension frontend for visited URL and webpage content analysis.
This solution makes sure that users' data, both personal and browsing, are not stored anywhere, as it operates on lightweight inference, thus guaranteeing users' privacy. The proposed system embeds advanced ML capabilities within the browser environment and hence offers adaptive, fast, and user-friendly protection against evolving phishing threats. It forms a practical bridge between academic research and real-world cybersecurity applications, providing a scalable approach that protects users in everyday interactions over the web.
The proposed system uses DistilBERT LLM in the LLM component of the system; hence, it provides deep contextual interpretation for webpage content to identify sophisticated phishing attacks.},
        keywords = {},
        month = {November},
        }