Abstract— There is an increasing need for strong detection techniques due to the growing intricacy of phishing strategies. Using machine learning methods, this session explores the field of phishing website detection. Our method uses CatBoost's discriminative capacity to examine the features of phishing websites with the goal of improving accuracy and efficiency. The suggested approach reduces security risks in a time when cyber-attacks are a common occurrence by offering users a smooth interface to determine the legality of URLs. Through combining knowledge from several detection techniques, such as similarity-based, list-based, and machine learning-based methods, we want to strengthen the defense against phishing attempts that are always evolving. By conducting a comprehensive analysis of the existing state of affairs and identifying areas that require further investigation, our project aims to enhance the field of phishing detection and protect people and organizations from financial losses and privacy violations.