inputAction

The action to take when harmful content is detected in the input. Supported values include:

• BLOCK – Block the content and replace it with blocked messaging.

• NONE – Take no action but return detection information in the trace response.