When I chat with the agent manually, it works perfectly. It retrieves data from Fabric correctly.
When I try to use the built-in Evaluation feature, it spins/loads for 10-15 minutes and finally returns an empty string: Response from assistant: ""
I also tried using the Copilot Studio Kit for testing, but it fails in the exact same way.
It seems like the evaluation runtime cannot handle the Fabric connection, while the standard chat runtime can.
Has anyone else encountered this? Is the Evaluation feature currently supported for Fabric Data Agents in Preview?
Thanks for your help.


Report
All responses (
Answers (