krishshirole/Pdf-RAG-Assistant ? reverse-engineered prompt

Build me a simple PDF question answering app where I can upload one or more PDF files, choose which ones I want to search, and then ask questions in plain English about their contents.

I want the app to extract text from the PDFs, split it into chunks, store it in a local vector database, and use retrieval augmented generation to answer questions with an LLM through OpenRouter. Please make it beginner friendly and keep the code easy to understand, since this is meant to show how a basic RAG system works.

Use a clean Streamlit interface so I can upload files, select PDFs, build the knowledge base, type a question, and see the answer along with the source PDF and page number. Keep the API key in a .env file and make sure the app can run locally with a simple command. If anything needs current setup details, look up the latest docs online and wire it up in the simplest working way.