When do you build a zap vs when do you build an agent?
I’m in the early, early phase of building an agent for some tasks and I’m feeling like it would make more sense to use a zap for things that don’t require context or interpretation. I could easily see myself building an agent and going overboard or over complicating something by using an agent. What are some of the questions or “tests” do you use to determine if you need a zap or an agent?