Human-in-the-loop
Human-in-the-loop describes an AI system that pauses to ask a real person for a decision at specific points, instead of running fully on its own.
Full autonomy sounds appealing until an agent hits a decision it genuinely shouldn't make alone — spending money, deleting something, or a judgment call with real consequences. Human-in-the-loop is the design pattern for building that pause in deliberately, rather than either blocking entirely or barreling ahead.
Assistant
An agent's turn in Masterforce ends one of two ways: it finishes and moves on, or it pauses and pulls in a live person when it hits something only a human can answer right now — the same underlying Approval primitive used for governance changes and other irreversible actions across the product.