Hello team:
For the evaluation mentioned in this video: Agent Skills with Anthropic - DeepLearning.AI, I still don’t quite understand about how that json file serves the purpose of unit test, may I know more details about how that unit tests run? Thanks!!