In this study, we present a conceptual design for a web privacy assessment tool, which is a browser extension promoting privacy and fair data practices. The tool gives the user a near real-time view on network traffic and data processing of a website. The solution makes use of large language models (LLMs) to analyze network traffic sent to third parties like analytics services and advertisers in the web environment. It detects leaks of identifying and contextual personal data, such as details related to the user’s health. The tool assesses the transparency of the privacy policy document provided on the website in order to see whether the user is adequately informed about the data shared with third parties. The tool also has the ability to detect dark patterns on cookie consent banners. Our solution is targeted for end users, and it is meant to be an intuitive and usable tool providing clear and timely information about the occurring data leaks. The novelty of the proposed solution lies in combining several features. It aims to offer near real-time notifications for users, puts emphasis on the sensitive contextual data leaks, has the ability to detect discrepancies between the actual network traffic and the privacy policy using AI, and provides a concise summary of data leaks detected on the analyzed website, as well as an assessment of the privacy and fairness of data processing practices. Our solution is informational rather than preventive by design. It increases the transparency of data processing and supports the user’s own decision-making when it comes to data protection.