Can You Prompt an LLM to Solve a Simple Puzzle Clearly?

I've observed something interesting: my toddler can quickly and somewhat effortlessly solve a simple physical puzzle, yet when I try to prompt large language models (LLMs) to guide me clearly through the solution, they struggle.

Even with clear photographs and detailed instructions, LLMs tend to give general advice or indirect methods rather than precise, actionable steps that clearly demonstrate correctness.

Have you tried something similar? Have you successfully "prompt-engineered" an LLM into giving clear, precise, step-by-step solutions for physical puzzles? If yes, what's your approach?