### Answer --- The behavior you're describing—where the LLM sends all fields including nulls in a tool call—is controlled by how you define your custom function/tool and how you instruct the LLM in your prompt. The available documentation doesn't provide a built-in Retell AI feature to automatically filter out null or unspecified fields from tool call arguments. Recommended approaches: 1. Prompt Engineering: In your agent's prompt or tool description, explicitly instruct the LLM to only include fields the user mentioned. For example, in your tool's

description

, you could add: "Only include fields the user explicitly provided. Do not include fields with null or empty values." 2. Handle on Your Server: When your custom function endpoint receives the tool call, filter out any null or empty fields before making your PATCH request to your lead system. This is a common pattern when the LLM may send more fields than necessary. 3. Tool Call Strict Mode: You can enable

tool_call_strict_mode

in your Retell LLM configuration, which ensures responses follow your provided JSON Schema more strictly. [(1) ]() I don't have documentation showing a native Retell AI setting that automatically omits null fields from tool call arguments. If you'd like, I can try to provide a general answer on how to structure your tool description or server-side logic to handle this. Check out these potentially useful sources: - [Update Retell LLM]() - [LLM WebSocket]() - [Update Conversation Flow]()

​ To learn more, see the sources I considered: [[1] Update Retell LLM](https://docs.retellai.com/api-references/update-retell-llm) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

Appreciate the feedback! Feel free to ask me again with additional guidance or context.