Mango is a widely cultivated fruit in tropical and subtropical regions, but is highly susceptible to various leaf diseases, which can reduce yield and fruit quality if not detected and managed early. Conventional approaches to disease diagnosis depend on manual examination, which is both time-intensive and susceptible to human errors. This research proposes an automated system for detecting and classifying mango leaf diseases with the integration of drone and a mobile application. The drone and mobile application both help capture images from the farm. The processing power constraint of the edge devices led to use of cloud server for processing images using a modified Convolutional Neural Network architecture which was designed and trained to classify mango leaves into healthy and diseased categories, achieving an overall classification accuracy of 98.63%.The cloud database is used for storing the obtained results which are then fetched by the mobile application for showcasing the real-time insights.The architecture ensures low latency and proper scalability by leveraging cloud computing.