Hey everyone! I’m new here and working on integrating RAG (Retrieval-Augmented Generation), and I’ve got a question. I’ve set up RAG with a vector database, so when a customer asks something, I can pull up multiple possible result.After that i create a tool called “anyQuestions.” this tools triggers in whenever a customer asks something, sends the data to my server, calculates the probability of each answer, and feeds it into the prompt. But for more specific questions, like “Can you book an appointment?”, I’ve got a different tool that handles those directly, so they skip the 'anyQuestions' process.Sometimes skip to appointmentBuilder I’m also using a Server URL as a webhook to capture conversation data. But so far, it’s not picking up anything on my end. Any tips on getting this to work—other than building a custom LLM? Thanks

Hey @Deniz Would you mind sharing the call id and by "But so far, it’s not picking up anything on my end" do you meant to say it's not sending server events to your server url? If not then do correct me.

You can close i create my custome llm to integrate thanks