DETE-X, a novel web application that aims to provide a robust system for sentiment-based cyber harassment detection and social media monitoring on the platform X. The application uses a fine-tuned XLM-RoBERTa model for deep contextual understanding, which then feeds into a Support Vector Machine (SVM) classifier to accurately detect and assess the severity of cyberbullying content in both English and Tagalog tweets. Once detected, the system provides on-demand alerts and actionable moderation suggestions to users, fostering safer online interactions.