This makes me think how AI turns SW development upside down. In traditonal development we write code which is the answer to our problems. With AI we write questions and get the answers. Neither is easy, finding the correct questions can be a lot fo work, whereas if you have some existing code you already have the answers, but you may not have the questions (= "specs") written down anywhere, at least not very well, typically.
At least in my experience the AI agents work best when you give them a description of a concrete code change like "Write a function which does this here" rather than vague product ideas like "The user wants this problem solved". But coming up with the prompts for an exact code change is often harder than writing the code.