How to Use a Binary Do Conditioning Variable to Simulate a Deliberate Action

So at a basic level, would I say I just introduced a second conditioning variable, like a binary do variable and I'm conditioning on that additional zero one do variable? Would that be something that would maybe be useful for say like, say it's like a reinforcement learning agent. And so when you can do that then causal inference is easy right all you do is just run the experiment. You know you don't have to do any adjusting for anything because you have access to the world model essentially and you can you can just run the Experiment.