FeedDiscussion
Alessandro Pieraccini
The agent-state-as-first-class-concept point maps almost exactly onto a state machine, planning, executing, waiting-for-approval, completed, needs-intervention are states with explicit valid transitions, not just a status string you branch on. Something like XState models this well since it makes illegal transitions (like completing from idle without executing) impossible by construction, instead of relying on every consumer of the state to check it correctly.
Hi, and thank you for this great article. The passage about the risk of turning every action into a confirmation is correct and particularly challenging. If the agent has to ask permission for everything, the system does not differ in any way from a “click-by-click” model, and maybe it would not even make sense.
Equally complex is understanding how to design thresholds, rollback and history well, therefore thinking not only about what AI can do, but also about how to make it possible to recover when it makes a mistake.