Each iteration assembles current state, available tools, prior observations, and constraints into the model input. The runtime executes valid actions and stops on completion, limits, failure, or human intervention.
An agent loop repeatedly asks a model to choose the next action until it reaches a goal or stopping condition.
Each iteration assembles current state, available tools, prior observations, and constraints into the model input. The runtime executes valid actions and stops on completion, limits, failure, or human intervention.