Plant diseases cause significant loss to both farmers and gardeners, due to various climate change factors like shifts in weather patterns which make them susceptible to various diseases and also lack of reliable tools and resources to identify diseases effectively. Another important factor is fragmented knowledge, due to which getting proper guidance becomes very tedious. Therefore, the proposed plant care system is a solution to these as the first module of it which deals with plant disease detection makes use of convolutional neural networks (CNN) which detects plant disease using the leaf images and the second module which consist of RAG and LLM based chatbot provides user guidance as to cure of such diseases and also measures as to how to prevent them. The CNN based plant disease detection model process and analyses leaf images such that the accuracy increases with training and in order to prevent overfitting, early stopping was practiced which stops training when validation accuracy no longer shows any improvement. This approach results in a training accuracy of 95% and a validation accuracy of 92%, ensuring reliable and effective identification of various plant diseases. Complementing this, the second module the chatbot, powered by LLAMA-2 and an efficient retrieval mechanism using chunking, embeddings, and FAISS, provides fast, accurate, and context-aware responses. Integrated into a plant care system, it can answer user queries about plant diseases, care instructions, and maintenance tips with natural, conversational interactions while easily updating its knowledge base to include new plant-related information.