Access to legal information in India is still a challenge to the average citizen. The factors contributing to this include the use of difficult legal jargon, the large length of the official legal texts, and the lack of cheap legal advice. The currently available legal chatbots and digital tools mostly depend on large language models only, which results in hallucinations, lack of explainability, and unreliable legal interpretations. LegalEase presented this project is a response to these constraints. It is an AI, powered conversational assistant based on a Retrieval, Augmented Generation (RAG) framework that provides legal guidance that is grounded, transparent, and in simple terms. The system takes statutory documents like the Indian Penal Code (IPC) and the Constitution of India and prepares them for use by extracting structured text, chunking, and semantic embedding using Sentence Transformer models. A FAISS, based vector retrieval method finds the most relevant legal clauses that are then combined with user queries to generate context, aware responses using a constrained language model. The experiments were run with labeled legal queries to assess the performance of the retrieval part separately from the generation part. The system showed high accuracy, precision, recall, and F1, score in retrieval, thus it was able to reliably identify relevant legal provisions and did so with very few unsupported responses. The major point of this work is the provision of a legal assistance framework that is explainable, retrieval, grounded, and thus contributes to raising legal awareness, lowering the risk of misinformation, and democratizing access to reliable legal knowledge for non, expert users.