If multiple component options are provided a "judge" AI will evaluate all the functional responses and will rank order them with an explanation of why they were each ranked that way. Rankings are not absolute, they are relative to the

of components contained within an array of ComponentOption types.

The AI's explanation of why it ranked the component the way it did. This is useful for understanding the AI's reasoning and can be used to improve future components as well as provide context to the user about why a particular component was chosen as the best option.

Optional, name of the option the AI gave it

Full details of the generated component option including functional, technical, code, dependencies and libraries

The user's provided feedback on the component option. Unlike the AIRank, this is a subjective rating provided by the user and is a number between 1 and 10, where 1 is the lowest rating and 10 is the highest rating.

If the host application provides a way for the user to provide feedback on the component option, this is the explanation of why the user rated the component the way they did if they provided feedback.