ETCHARLAREVANTHRAO/GRAG ? reverse-engineered prompt

Build me a simple demo that shows how graph based retrieval and answer generation work together.

I want a clean app where I can ask a question, load a small sample knowledge graph, and see how the system finds related nodes, narrows down the useful subgraph, and then generates an answer from that context. It should make the whole flow easy to understand, with a clear visual view of the graph, the retrieved subgraph, and the final response side by side.

Please include a few example questions that need multi hop reasoning, and make the output feel accurate and grounded in the graph instead of guessing. If you need to look up current docs or best practices online, go ahead. Keep it simple, polished, and easy to run locally.