Phishing attacks are a prevalent threat to individuals and organizations, resulting in huge financial losses, data breaches, and reputational losses. The current research suggests a new Email Monitoring System specifically designed for phishing threat detection using cutting-edge machine learning (ML) and natural language processing (NLP) methodologies. The system analyzes the content and metadata of emails for identifying suspicious patterns typical of phishing campaigns. A vast corpus of authentic and phishing emails was used for training diverse machine learning models such as decision trees, support vector machines, and neural networks. The system also includes a quarantine folder where incoming emails suspected to be threats are temporarily kept for users to check at their convenience. The outcomes reveal that the suggested system remarkably improves cybersecurity efforts by offering real-time detection and alerts, thus protecting sensitive information and ensuring security to all concerned stakeholders, including organizations. The outcomes also reveal that the Email Monitoring System outperforms available solutions in terms of accuracy as well as flexibility to adapt to changing phishing attack strategies.