This paper proposes a novel system for pest detection and classification of agricultural pests using the YOLOv8 and MobileNet architectures to improve the management of pests in agriculture. YOLOv8 provides accurate, real-time detection of objects, while MobileNet, an efficient convolutional neural network, has a 95.99% accuracy rate for classifying pests, specifically designed for devices with fewer capabilities. Through the combination of these deep learning techniques, our model offers excellent performance, scalability, and farmers in different agricultural environments. After pest identification, the system employs web scraping to provide users with useful advice and preventive action, making it more useful. Thorough testing confirms the strength and efficiency of the system in usefulness in real-world scenarios, demonstrating its potential to revolutionize pest detection and control. Through simplifying and speeding up pest identification, our system empowers farmers with accurate information for timely decision-making and early interventions, enhancing pest control methods. This innovative technology enables sustainable crop production, increasing yields and global food security. In short, our project innovates agricultural technology by offering a low-cost, scalable solution to pest management issues, opening the door to more sustainable farming practices. The integration of real-time detection and effective classification is a major leap forward in sustainable agriculture and increased global food security.