That might be the case but it's definitely not watching what you're doing, like you suggested. There may be prompts that are triggered by repeated actions (think hitting the Shift key 5 times in Windows and getting the Sticky Keys prompt) but there's nothing in there that's watching you and suggesting where you can make improvements.

Again, what you're asking for isn't a trivial implementation.

What I'm asking for is pretty much exactly what JetBrains have done in IntelliJ. When they system sees me do something manually that could be done with voice, tell me. For example, if I enter an address into maps then start navigation, the machine could suggest I use Siri next time with the phrase "navigate to 123 main street".

At the very least, it would be nice if I could say "Hey Siri, what voice commands can I use with app X?"