Overview
LiveKit Agents includes a number of ready-made tools components that you can use to reliably solve recurring problems in voice AI. These tools have been tested and tuned based on real-world scenarios and can help you rapidly build your own voice AI applications.
Tools let your agent perform actions and call external logic in response to the conversation. In addition to the prebuilt tools, you can define your own function tools or use other tool types. For more information, see Tool definition & use.
These tools can be found in the livekit.agents.beta.tools module.
LiveKit Agents for Python contains the following prebuilt tools:
| Tool | Description |
|---|---|
| EndCallTool | Gracefully end the call and disconnect from the room. Optionally deletes the room and runs custom goodbye instructions. |
| send_dtmf_events | Send DTMF tones to telephony providers. For navigating IVR menus and phone systems. |
Usage
Add the prebuilt tool to your agent's tools list when constructing the agent. The LLM can then call it during the conversation when appropriate. See each tool's page for specific usage instructions.
Customization
Prebuilt tools support customization so you can adapt them to your use case. Configure behavior via constructor parameters such as extra_description or end_instructions. See each tool's parameters section for details.