Sebastián Estévez

The double quotes standard around upper case keyspaces and columns is a cassandra standard. You have to do it in cqlsh, at the driver, etc. I see your point about...

With JSON, this becomes a bigger issue because now we're requiring the column names to come escaped in the JSON file. Nobody will have this. I think it's time to...

Here's a new PR to deal with the JSON issue.

@VRSEN I fixed your bug here I hope that unblocks you: https://github.com/datastax/astra-assistants-api/issues/21

I was testing further and noticed this issue https://github.com/datastax/astra-assistants-api/issues/40. Right now agency-swarm seems to work great without file-search. Switching to prefixed IDs should fix retrieval.

Fixed the IDs and added a working example here https://github.com/datastax/astra-assistants-api/blob/main/examples/python/agency-swarm/basic.py

Added unit test [and uncovered another bug] Thanks!

I'm hitting `RuntimeError('dictionary changed size during iteration')` in some cases, is wrapping > for k in passed_params.keys(): is `list()` the right fix? It gets me past the error but eventually...

Hmm this situation reproduces on `main` with the existing cohere unit test -- test_cohere_completion: ``` test_cohere_completion.py:20 (test_chat_completion_cohere) def test_chat_completion_cohere(): try: litellm.set_verbose = True messages = [ { "role": "user", "content":...

> passed_params = locals().copy() seems to fix it but `locals()` is used all over the code base, wondering if this is going to lead to similar issues elsewhere.