Timely and accurate classification of plant diseases is essential for maintaining agricultural productivity and food security. Conventional disease detection involves manual inspection, which is expertise-based and time-consuming. Here, we exploit the strength of deep transfer learning for enhancing plant disease classification using the YOLOv8 model. We fine-tune a pre-trained YOLOv8 model (yolov8n-cls.pt) on a well-prepared dataset of plant disease images with a training schedule of 10 epochs with 224  \(\times \)  224 pixel image resolution and batch size of 16. The model performs well with an accuracy of 95.67%, precision of 95.70%, recall of 95.67%, and F1-score of 95.66%. The results demonstrate the ability of the model to detect pathogen-based diseases accurately even with limited labeled data. The high performance metrics reflect the promise of YOLOv8 as a reliable tool for real-time disease detection in intelligent agricultural support systems. This work reinforces the efficacy of deep transfer learning in addressing key challenges in plant disease management and hence facilitating scalable, efficient, and reliable solutions in modern agriculture.