akamai-developers/rag-langgraph-k8s-quickstart ? reverse-engineered prompt

Build me an airline policy chatbot that can answer questions like baggage rules, change fees, cancellations, and other travel policies from uploaded airline documents.

I want a simple web chat app where a user asks a question and gets a clear answer with the relevant sources behind it, not just a generic AI response. The app should read policy files like PDF, HTML, and text files, turn them into searchable knowledge, and use that to answer in a conversational way while remembering the chat history. If the answer is not in the documents, it should say so honestly.

Please make it work as a full project I can run and deploy, including the backend, the document ingestion flow, and the infrastructure setup for hosting it on Kubernetes in Akamai Cloud. Use current docs online if you need to. Also include the setup for storing embeddings and chat state in PostgreSQL, and make it easy to load new airline documents later.