I'd love to help you with making the agent LLM stop processing once it enters a function tool and simply display a message. Here are a couple of friendly approaches you can try: • Return No String: If your function tool doesn’t return a string (or returns

None

), the LLM won’t generate a follow-up response. For example:

@function_tool
async def example_tool(self):
    self.session.say("Processing your request.")
    # No return value prevents further LLM response

• Use `ev.cancel_tool_reply()`: When multiple tools run in parallel, you can listen for the

function_tools_executed

event and call

ev.cancel_tool_reply()

to stop any LLM-generated reply. If you’d like, I’m here to walk you through these methods or provide more details—just let me know! Sources: agent.py | tool_calling/update_tools.py | Agent speech and audio | LiveKit Docs | examples/voice_agents/session_close_callback.py