Hello Community,
My agent is fabricating a dataset even though I have specified a data explore and provided instructions for preventing the agent from inventing information. I have tried following suggestions here for Vertex AI How to make sure the agent answers only based on the data from the tool but I’m using the conversational analytics capability in the Looker UI.
Ive tried honing my agent’s instructions using VertexAI, but not sure what I can do next. Any suggestions/tips?
As far as I’m aware, you can only delete personal folders via the API - you cannot do this in the web UI.
I deal with this by following an offboarding process:
1. If an employee is being offboarded, I check whether they have any content saved in their personal folder.
2. If they do, and it has not been viewed by anyone else in the last 90 days, then I delete it.
I have a dashboard based on the System Activity Explorers that quickly gives me this information when I filter it for the specific user. I appreciate this manual process would be unacceptable if you have a very large instance, with 1000’s of users.
@Looker Staff: the above post by krokrodil07 is a direct copy of one I posted HERE. I assume it has been posted by a bot. There does not seem to be an option to report it as abuse.