Using relevant diagnostic measurements such as systolic blood pressure, diastolic blood pressure, fasting blood sugar level, hemoglobin A1c level, serum creatinine level, blood urea nitrogen level, glomerular filtration rate, protein levels in urine, albumin-to-creatinine ratio, serum sodium level, serum potassium level, serum calcium level, serum phosphorus level, hemoglobin level, total cholesterol level, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglycerides level, as well as lifestyle and environmental factors, suitably complex artificial intelligence (AI) models generated based on suggestions extracted from generative AI tools such as large language models (LLMs) through prompt engineering are trained and deployed to automatically diagnose chronic kidney disease for clinical decision support purposes. The results indicate that the application of prompt engineering to generative AI tools coupled with AI expertise is a viable approach to the development of AI models for automatic chronic kidney disease diagnosis on the basis of diagnostic measurements, lifestyle and environmental factors. The trained AI models could be incorporated into a modular comprehensive AI-driven healthcare system designed to provide actionable insights that can support clinical decision-making practice.