Hi @ocejudo2000 !! Welcome to our community 
I was just playing around with flowise recently.
As it uses langchain under the hood, it calls index
what we call collections
(or previously class/schema).
On that case, you can fill any value for in that index field. If that index exists, Flowise will insert your objects. Otherwise, it will be created.
for host and credentials, those informations are available in the details of your cluster in WCS console: https://console.weaviate.cloud/
There are some pitfalls I have faced while using flowise with Weaviate. 
First, you will only be able to do a similarity search in Weaviate when you provide the vector for the query. So no near_text, image, generative etc for those collections created in Flowise will not work.
This happens because langchain/flowise takes care of the collection creation and the vectorization for you, and when it creates the collection, it never specifies the vectorizer or generative module to be used in Weaviate.
This can be avoided by creating the collection (or index) beforehand and explicitly specifying the modules to be used. I have crafted a recipe here that shows how to do that:
The second pitfall I have seen so far is that both flowise and langchain implementation doesn’t support multi tenancy, so if you are planning on supporting multiple clients, you will naturally create one index per customer.
That works, but will not scale well if needed (think hundreds of thousands tenants.) as it may affect Weaviate startup time.
We have been recently working on new developments for langchain in both python and typescript, so hopefully we can close those pitfalls down the road.
Let me know if that helps 