To me this reads like trying to fit a solution we know to a problem we haven't yet defined.

The problem is more around ops / visibility / delegation / orchestration of agents, but the solution is being misslabelled as "IDE" which I feel like is the wrong analogy although the right "in-between" step towards what the next thing will be.