would like to know if we can set the name of agent that has been created for explicit dispatch. I tried several ways but could never set its name.

Thanks for reaching out, Suba. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

I am using python

Since you're using Python, you can easily set the agent name by specifying the

agent_name

field in the

WorkerOptions

. Here's a simple example to guide you:

opts = WorkerOptions(
    entrypoint_fnc=entrypoint,
    agent_name="test-agent"
)

Just replace

"test-agent"

with the name you'd like for your agent. If you have any questions or need a hand with this, feel free to reach out—I'm here to help!

so will this reflect in the UI?

Hi there! I wanted to share a little insight about the

agent_name

set in

WorkerOptions

. It doesn’t show up directly in the LiveKit UI as the agent's visible name; instead, it’s used behind the scenes for dispatch control and routing. If you'd like the agent’s name to be visible in the UI, you’ll want to set it explicitly when the agent joins a room. For example, when accepting a job, you can do this:

await req.accept(name="Call Agent")

The

name

parameter here is what determines the participant’s visible name in the room. If you have any questions or want me to walk you through it further, just let me know—I’m here to help! Sources: Worker options | LiveKit Docs | Agents telephony integration | LiveKit Docs

no I meant the dashboard of livekit

I wanted to share a little insight about the

agent_name

set in

WorkerOptions

. It’s mainly used for explicit dispatch and is accessible internally or through the API. That said, I’m not entirely sure if this name shows up in the LiveKit dashboard UI, so it might be a good idea to take a quick look there to confirm. If you have any more questions or if there’s anything else you’d like help with, please don’t hesitate to reach out—I’m here for you! Sources: Dispatching agents | LiveKit Docs | livekit-agents/livekit/agents/worker.py

I have configured the name and it's getting displayed in live kit UI successfully. however when I mention the agent name in sandbox option for explicit dispatch, it always sends the other agent in the project

I'm here to help you resolve the issue with the wrong agent being dispatched, even after specifying the

agent_name

. Let's go through a few important points together: 1. Exact Match: Please make sure the

agent_name

set in

WorkerOptions

exactly matches the name used in your dispatch request. Remember, agent names are case-sensitive, so even small differences matter. 2. Unique Agent: It's important that only one agent worker is registered with the specified

agent_name

. If multiple agents share the same name, the system might dispatch any one of them, which could cause confusion. 3. Correct Dispatch Method: Double-check that you're using the right dispatch method, like API or CLI, and that the

agent_name

is specified properly. For example, in Python, it looks like this:

dispatch = await lkapi.agent_dispatch.create_dispatch(
    api.CreateAgentDispatchRequest(
        agent_name="test-agent",
        room="my-room"
    )
)

1. Automatic Dispatch Disabled: Please verify that automatic dispatch is turned off for the agent to avoid conflicts. 2. Room Name: Ensure the room name in your dispatch request matches exactly the room you intend to use. 3. Check for Typos: It’s always good to double-check there are no typos or case mismatches in the

agent_name

. If you continue to experience this issue, it might help to check the running agent instances and logs to see which agents are registered and receiving dispatch requests. Feel free to reach out if you want me to guide you through those steps or if there’s anything else I can assist you with! Sources: Agent dispatch | LiveKit Docs | livekit-api/livekit/api/agent_dispatch_service.py

how to check below point? 1. Automatic Dispatch Disabled: Please verify that automatic dispatch is turned off for the agent to avoid conflicts.

Hi there! To check if automatic dispatch is turned off for an agent, you can take a look at the agent's

WorkerOptions

configuration. When the

agent_name

field is set, it means automatic dispatch is disabled. Here’s a quick example in Python:

opts = WorkerOptions(
    entrypoint_fnc=entrypoint,
    agent_name="my-telephony-agent"
)

So, if you see the

agent_name

set, automatic dispatch won’t happen automatically. Instead, the agent will need to be dispatched manually through the API, SIP rules, or participant tokens. If you have any questions or want me to walk you through it, just let me know—I’m here to help! Sources: Agent dispatch | LiveKit Docs | Agents telephony integration | LiveKit Docs | Dispatching agents | LiveKit Docs

I named the agent. but get below in logs when starting a call: {"message": "failed to connect to livekit, retrying in 0s", "level": "WARNING", "name": "livekit.agents", "exc_info": "Traceback (most recent call last):\n File \"/app/.venv/lib/python3.13/site-packages/livekit/agents/worker.py\", line 689, in _connection_task\n await self._run_ws(ws)\n File \"/app/.venv/lib/python3.13/site-packages/livekit/agents/worker.py\", line 770, in _run_ws\n await asyncio.gather(*tasks)\n File \"/app/.venv/lib/python3.13/site-packages/livekit/agents/worker.py\", line 742, in _recv_task\n raise Exception(\"worker connection closed unexpectedly\")\nException: worker connection closed unexpectedly", "timestamp": "2025-09-20T234632.475636+00:00"}

I'm sorry to hear you're encountering the errors "failed to connect to livekit, retrying" and "worker connection closed unexpectedly." These usually happen because of connectivity, authentication, or server-side issues. Let’s work through some steps together to help you resolve this: 1. Check Authentication: Please double-check that the API key and secret you're using for the agent are correct and still valid. 2. Verify WebSocket URL: Make sure the WebSocket URL in your configuration is accurate and properly set. 3. Network Connectivity: It’s a good idea to look into any network issues that might be affecting the connection, like firewalls or VPNs that could be blocking access. 4. Server Logs: Take a look at the LiveKit server logs for errors such as

HandshakeAgentWorker

or

DispatchAgentWorkerSignal

. These can give clues about authentication or connection hiccups. 5. Connection Test: Try using the Connection Test utility to check your WebSocket, WebRTC, and TURN server connectivity to the LiveKit servers. 6. Worker Load: Also, ensure the worker isn’t overloaded and is operating within the configured thresholds. If you’d like, I’m here to help you through any of these steps or answer any questions you might have. Just let me know! Sources: livekit-protocol/src/livekit.rs | protobufs/livekit_models.proto | livekit-rtc/livekit/rtc/_proto/participant_pb2.pyi | pkg/service/agentservice.go | livekit-agents/livekit/agents/_exceptions.py

the problem is, the agent is not joining room if the name is set in worker options. I have set the agent name in sandbox options for explicit dispatch.

I'd love to help you get this sorted out! Could you please share a bit more information with me? • Are you seeing any specific error messages or logs when the agent has trouble joining the room? • Have you checked that the

agent_name

in

WorkerOptions

exactly matches the name used in the sandbox options for explicit dispatch? • Is the room name entered correctly in the dispatch request? Having these details will really help us figure out what's going on together.

• Is the room name entered correctly in the dispatch request? how to check this?

Let's make sure the room name is correctly specified in your dispatch request. You can easily check this by using the

listDispatch

method from the

AgentDispatchClient

. This handy method will fetch all dispatches for a particular room and help confirm if the dispatch was created with the right room name. Here's a simple example in TypeScript to guide you:

const agentDispatchClient = new AgentDispatchClient(
  process.env.LIVEKIT_URL,
  process.env.LIVEKIT_API_KEY,
  process.env.LIVEKIT_API_SECRET
);

const roomName = 'my-room';
const dispatches = await agentDispatchClient.listDispatch(roomName);
console.log(`There are ${dispatches.length} dispatches in ${roomName}`);

If the room name is spot on, you’ll see a list of dispatches returned. Feel free to reach out if you have any questions or need a hand with this! Sources: Agents telephony integration | LiveKit Docs | packages/livekit-server-sdk/src/AgentDispatchClient.ts | Dispatching agents | LiveKit Docs | Agent dispatch | LiveKit Docs

I am using the default sandbox in livekit. below is the agent code for session creation in agent.py. should we mention anything here ? await session.start( agent=Assistant(), room=ctx.room, _room_input_options_=RoomInputOptions( # LiveKit Cloud enhanced noise cancellation # - If self-hosting, omit this parameter # - For telephony applications, use

BVCTelephony

for best results _noise_cancellation_=noise_cancellation.BVC(), ), )

In the default sandbox, agents are automatically dispatched unless you set an

agent_name

in

WorkerOptions

. If you'd like to use explicit dispatch, here’s what you can do: 1. Set the

agent_name

in

WorkerOptions

to disable automatic dispatch. For example:

cli.run_app(agents.WorkerOptions(
    entrypoint_fnc=entrypoint,
    agent_name="my-agent"
))

1. Then, use an explicit dispatch method, like the

AgentDispatchService

API, to assign the agent to a room. For example:

await lkapi.agent_dispatch.create_dispatch(
    api.CreateAgentDispatchRequest(
        agent_name="my-agent",
        room="my-room"
    )
)

Also, just a heads-up: the

session.start

method in

agent.py

kicks off the agent's session after dispatch—it doesn’t handle the dispatch itself. So, please make sure the room name (

ctx.room

) is correctly passed to

session.start

. If anything’s unclear or you’d like a hand with this, feel free to reach out—I’m here to help! Sources: Authentication | LiveKit Docs | agent.py | Agent dispatch | LiveKit Docs | Dispatching agents | LiveKit Docs | Agents telephony integration | LiveKit Docs

where should I add this code? await lkapi.agent_dispatch.create_dispatch( api.CreateAgentDispatchRequest( agent_name="my-agent", room="my-room" ) )

I'd love to help you out! Could you please share a bit more about where exactly in your application flow you'd like the agent to be dispatched? For instance: • Would you want the agent to be dispatched when a specific event happens, like when a user joins a room or a call starts? • Are there any triggers or conditions you're using to decide when the agent should join a room? Knowing these details will really help us find the perfect spot to add the

create_dispatch

API call for you.

I am using the default sandbox in live kit. want the specific agent to be dispatched when a user joins the room

When a user joins a room in the default sandbox, you can easily dispatch a specific agent by adding the

create_dispatch

API call right where your application handles the user joining the room. This way, the agent will be sent directly to that room, making the process smooth and clear. Here’s a simple example of how you might do this in the event handler or function that deals with a user joining:

async def on_user_join(room_name):
    await lkapi.agent_dispatch.create_dispatch(
        api.CreateAgentDispatchRequest(
            agent_name="my-agent",
            room=room_name
        )
    )
    print(f"Agent 'my-agent' dispatched to room {room_name}")

Just swap out

on_user_join

with the actual function or event handler you’re using for user room joins. If you have any questions or want to walk through this together, I’m here to help!

I only have sandbox, no other application

I understand you're working with the default sandbox in LiveKit, and I’m here to help you with explicit dispatch. You can still achieve this by manually triggering the

create_dispatch

API call. Since there’s no external application involved, you can simply run this API call through a script or tool whenever you want to dispatch an agent to a room. Here’s a friendly example in TypeScript to guide you:

const agentDispatchClient = new AgentDispatchClient(
  process.env.LIVEKIT_URL,
  process.env.LIVEKIT_API_KEY,
  process.env.LIVEKIT_API_SECRET
);

const roomName = 'my-room';
const agentName = 'my-agent';

await agentDispatchClient.createDispatch(roomName, agentName);
console.log(`Agent ${agentName} dispatched to room ${roomName}`);

Feel free to run this script whenever you need to dispatch your agent. And of course, if you have any questions or need a hand with anything else, just let me know—I’m here to help!

can you send python code

Thanks for reaching out! Right now, the

create_dispatch

API call is definitely available in TypeScript. However, the documentation doesn’t clearly confirm if it’s directly supported in Python just yet. If you’re working with Python, a good approach might be to look at the TypeScript example and see if you can adapt it to your setup—especially if the Python SDK offers similar features. If you have any questions or need a hand with this, please don’t hesitate to ask. I’m here to help!