In the React-style paradigm, when does the agent stop iterating?

• A. After a fixed number of iterations.
• B. When the user manually stops it.
• C. When no tools remain available.
• D. When the goal is met.

Answer: (D)

In a React-style loop, the agent keeps going as long as it hasn’t achieved its objective. It repeatedly observes the current state, decides on the next action, executes it, and uses the result to update its understanding. The loop stops when the goal is met because once the desired outcome is reached, there’s no further improvement to pursue, so there’s nothing left to do. This makes goal fulfillment the natural stopping condition.

If you think about the other possibilities, they rely on external controls or constraints rather than the agent’s own progress: a fixed number of iterations imposes a preset budget, a manual stop depends on a user decision, and running out of tools is a resource constraint rather than the moment the objective has been achieved. The core idea is to continue until the goal is satisfied.