Using OpenAI vector stores in AI Library as RAG technique?

[tinsheep]

I want to use the Teams AI Library in my Custom Engine Agent and I see the examples that use Azure Cognitive Search. My SaaS app creates and uses Azure OpenAI Service Vector stores. I would like to be able to use them like I would use Azure Cognitive Search indexes in my Azure Open AI deployment? It looks like I can use them with assistants but is there a way to use the RAG technique to resister that Vector store with the planner as a data source and pass that configuration to the service to augment the prompt? Thanks!