The current system processes tool calls concurrently using separate asyncio tasks, and there are mechanisms to selectively ignore tool calls (such as when

tool_choice

is set to "none"). However, there is no built-in method to explicitly pause the processing of new tool calls while another is still running. To implement this functionality, you would need to: • Modify the tool execution loop to check for a pause condition before starting new tool calls. • Use a synchronization primitive (like an

asyncio.Event

or

Lock

) to control when new tool calls can be processed. • Extend the existing logic (such as

tool_choice

) to support dynamic pausing, not just static ignoring. The current architecture provides patterns for task management and coordination, but pausing tool call processing would require custom implementation based on these patterns. If you need more detailed guidance on implementing this, please specify your environment or share more about your use case. Sources: livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/llm.py