Chronic kidney disease (CKD) is a health condition that happens when the kidneys can no longer filter toxins effectively from the blood. Also known as renal failure, CKD is a progressive and irreversible condition managed mainly through hemodialysis or kidney transplantation. Traditional diagnosis methods are predominantly manual, which can be slow and susceptible to human error. Consequently, there is increasing interest in automated diagnostic systems powered by machine learning (ML), although many existing models still have performance limitations. This study focuses on developing a Decision Tree (DT) classifier for CKD diagnosis. Data was sourced from the UCI Machine Learning Repository, comprising 400 individuals 250 with CKD and 150 without the condition. The dataset includes 24 features, and feature selection was performed using chi-square method. The data was split into a 70:30 ratios for training and testing, and a DT model was built and evaluated using sensitivity analysis. The DT model achieved excellent performance, recording 97% accuracy, 93% precision, 97% recall, and a 95% F1-measure. These results demonstrate strong predictive capability and indicate that the model is highly promising for automated CKD detection. Future research should explore deploying ML models for real-time clinical diagnosis.