Phishing attempts, which mimic trustworthy websites in order to get private information like login credentials and financial data, continue to be a major concern to internet users. Because these threats are dynamic, early and precise detection is crucial to enhancing cybersecurity. A feature-driven method for detecting phishing URLs by examining their structural and domain-related attributes is presented in this paper. The method makes use of a dataset of labeled URL occurrences classified as phishing, suspicious, and legitimate that was taken from a publicly accessible Kaggle source. Key signs including URL length, the inclusion of strange symbols, the use of secure protocols, and domain-specific attributes are captured using a thorough feature extraction procedure. An efficient classification system that can differentiate between dangerous and benign URLs is created by processing these properties. The suggested system's performance is assessed using common metrics, showing that it can reliably detect phishing attempts while preserving a balanced detection rate. Furthermore, the system is practical for real-world applications since it is integrated into an intuitive web interface that facilitates real-time URL inspection. All things considered, the suggested method provides a dependable and scalable phishing detection solution; it can be improved by adding sophisticated data sources and adaptive learning methods.