In the era of precision medicine, Artificial Intelligence (AI) plays a pivotal role in medical imaging by enhancing diagnostic accuracy and efficiency. However, most existing AI models function as opaque “black boxes,” raising concerns regarding trust, transparency, adaptability, and clinical adoption. To address these challenges, MedXAgent, an intelligent agent–based framework designed for automated diagnosis of medical images, including chest X-rays and brain MRIs is proposed here. The framework integrates Explainable AI (XAI) techniques to provide transparent, human-interpretable visual explanations. Specifically, MedXAgent combines predictive outputs with visualization methods such as Grad-CAM and SHAP, which highlight salient image regions contributing to decision-making. The system is evaluated on public benchmark datasets, namely ChestX-ray14 and BraTS, demonstrating competitive diagnostic accuracy alongside interpretable outputs. Performance is assessed using standard quantitative metrics including AUC, IoU, and Dice coefficient, while explanation quality is evaluated through insertion and deletion scores. Experimental findings indicate that MedXAgent effectively balances diagnostic accuracy with explainability, thereby offering potential utility for research, clinical decision support, and deployment in real-world healthcare settings.