Great premise but can't really agree with the execution. Felt like this makes too many implicit assumptions about LLM capabilities and traps without differentiating enough between a smart human vs AI.