I have a workflow where the execution steps are generated by the AI in the first step of a workflow. Then based on context of the steps different agents execute the steps. Right now I have put all the generated steps into a single workflow step and running it in a for loop. Which works for now but I wanted to know if there is a better way to do it? The goal is to get more control for each generated step execution.
Current Workflow:
Generate Steps -> Execute Steps (for loop inside here) -> Validate -> Finish