Julian Valline

I noticed that in the API there is no mention how to indicate the allowed_types String. I found out that it needs to be indicated as a MIME type (e.g....

I noticed that in the API, there is no description about when_no_intent_matched()  Also, in the RAG agent example documentation page, it still shows a comment to fix the no_intent_matched...

Right now uploading Files through the UI for RAG can only be done on Agent level. Would be nice if we could implement this in the backend.

Would be nice if we could preprocess the user language from text or audio files as well

To consider if we want to add reply_speech function in TelegramPlatform. If not in this PR, open an issue to keep track of it _Originally posted by @mgv99 in https://github.com/BESSER-PEARL/BESSER-Agentic-Framework/pull/129#discussion_r2126989458_