“It’s the Architecture, Stupid” — Why Prompt Engineering Won’t Fix Agents

A developer argues that prompt engineering alone cannot fix AI agents, advocating for runtime state verification and action validity gating before execution. The approach, exemplified by DESi, ensures that answers and actions occur only when required conditions are satisfied, not merely because the model can generate them.

Yes — I agree. “Ask if unsure” must be a runtime state, not a prompt instruction. I think your “method validity gating” is the broader category. What I am trying to isolate is one specific case inside it: action validity before execution. Before an agent executes a skill, API call, code edit, or physical device action, the system should verify whether the instruction is complete enough to execute. If required inputs are missing, the state should become clarification, not inference. In that sense, DESi seems to gate whether a method is valid for producing an answer, while I gate whether an action is valid for execution. Both point to the same shift: answers and actions should not exist merely because the model can generate them. They should exist only when the required conditions are satisfied.