A Retrieval-Augmented Generation (RAG) system that answers questions about the AWS Customer Agreement using semantic search, embeddings, and LLM-powered answer generation.