Pamela Fox

Ah okay, I wasn't sure if deployment status API support was currently released. I can monitor the situation once that's released.

Thanks for filing, I also received that email. I've asked my colleagues if they have the Bicep handy for the necessary settings, if not we can figure it out.

I've run evaluations on pulling in more documents for the RAG flow, and the results are often not better, due to the increase in irrelevant documents. Therefore, I think a...

It might be good to check for secrets usage just to make sure there aren't any there that are using current secrets, since they result in security alerts. But agree...

Suggested for releasse notes- Breaking change, made to improve security: The `container-app.bicep` and `container-app-upsert.bicep` modules now mark the `secrets` parameter as `secure()`, and expect `secrets` to be an object (versus...

@weikanglim Ah yeah, my placeholders were misleading. Updated suggested text to look more like a real diff (such as https://github.com/Azure-Samples/langfuse-on-azure/commit/105148fffde072dd03e56c8d686963e665b8013e)

Is it possible you're logged in with a different account now? Can you check the existing roles in the portal to see what user they're associated with. A quick fix...

Those role assignments are declared in main.bicep. If you haven't modified them, then there should only be a single role with each name. You can try changing names there, however.

The position shouldn't matter, what matters is the enclosing square brackets, like [D0865021125133_CoA.pdf#page=1] The citation extracting code is in AnswerParser.tsx, so you can modify that to be looser if the...

@Daimler-Garay I've just added the actual allowed citations in the most recent version of the prompts- https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/approaches/prompts/chat_answer_question.prompty#L36 Have you tried that approach? (The frontend can then check citations against a...