Might become something remarkable soon. But right now it’s basically just multiple LLm prompt + responses in a row. More effective to have prompt + response between the LLM and a human at present