Is the response of tool calls (in my case that's text returned from n8n) directly used as the assistant answer? Or does the unterlying LLM create an answer based on the tool call response? Thanks so much for your help!

Based on the information I found, tool call responses in Vapi are typically processed by the LLM rather than being directly used as the assistant's answer. The LLM receives the tool call result and generates a new response based on it, as shown in this workflow:

1. LLM makes tool call
2. Tool (n8n) returns result
3. Result is packaged into follow-up API call
4. LLM generates new response incorporating the result

This allows the assistant to provide context and natural language around the raw data from your n8n integration. Sources: - [Custom LLM Tool Calling Integration](https://docs.vapi.ai/documentation/advanced-concepts/customization/custom-llm/custom-llm-tool-calling-integration) - [Discord discussion on tool call responses](https://discord.com/channels/1211482211119796234/1337465172263960637)

The tool call response is processed by the LLM, which generates a conversational answer based on the tool’s output, rather than using the tool’s response directly.

Ok, makes sense, thank you!