guardContent

The output content that's validated by the Automated Reasoning policy. This represents the foundation model response that will be checked for accuracy.