The primary aim of this research is to spread awareness about the existing cyber threats in today’s world. Phishing remains a significant cybersecurity threat, capitalizing on the widespread use of the Internet to deceive users and steal sensitive information. Traditional detection methods, which often rely on blacklisting, struggle to keep up with the rapidly evolving tactics used by attackers. This study proposes an advanced approach to phishing URL detection by employing machine learning techniques to identify malicious URLs. Utilizing a dataset of 3,000 labeled URLs (1,500 malicious and 1,500 benign), we extract key features indicative of phishing behavior, such as URL length, domain characteristics, and the presence of suspicious elements. We train and evaluate several machines learning models, including LightGBM, Random Forests, and XGBoost. Our results reveal that LightGBM achieves the highest accuracy, demonstrating superior performance in distinguishing between malicious and benign URLs while minimizing false positives. This approach highlights the potential of machine learning to enhance phishing detection and offers a robust defense mechanism against evolving cyberthreats.