Motif is a PyTorch-based project that trains AI agents to play NetHack by deriving a reward function from the preferences of a Large Language Model (LLM). It can generate behavior that aligns with human intuition and can be guided through prompt modifications.