How can i change the name of the agent participant that will be joining a livekit room. I have used the

await ctx.agent.set_name("Agent name")

But when two agents are joining the same room, they are being named "simulated_agent_y7236" by livekit itself. I dont have this issue when only a single agent is entering the room. Using v1.2.8 of agent and for front end i am using the meet git example from livekit itself

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

in the entrypoint function, this is what i do:

await session.start(agent=agent, room=ctx.room, room_input_options=RoomInputOptions(participant_kinds=[rtc.ParticipantKind.PARTICIPANT_KIND_STANDARD]))
    
    await ctx.connect()
    
    await ctx.agent.set_name("Candidate")

It looks like the issue with

ctx.agent.set_name

not working as expected when multiple agents join the room might be related to how

RoomInputOptions

is set up. By default, LiveKit filters participants by their kind, and agents are identified as

PARTICIPANT_KIND_AGENT

, not

PARTICIPANT_KIND_STANDARD

. This means agents might not see each other unless we explicitly configure it that way. To help fix this, you can update your

RoomInputOptions

to include

PARTICIPANT_KIND_AGENT

like this:

room_io = RoomIO(
    session,
    room=ctx.room,
    input_options=RoomInputOptions(
        participant_kinds=[rtc.ParticipantKind.PARTICIPANT_KIND_AGENT]
    )
)

Another approach some use is the

listen_to_agents=True

flag, but please note that this depends on the SDK version you’re working with. Once you have agents set up to detect each other, the

set_name

method should start working smoothly. If you run into any more questions or need a hand with this, feel free to reach out—I’m here to help! Sources: Runtime/Scripts/Participant.cs | livekit/src/room/participant/local_participant.rs | livekit-android-sdk/src/main/java/io/livekit/android/room/participant/Parti... | pkg/rtc/participant.go | How to Enable Agent-to-Agent Speech Detection in LiveKit

now what should i do with the roomio object after i created it, should i pass it to somewhere?

Once you create the

room_io

object, you can use it to manage the room's input and output streams. A great way to do this is by passing it to the session or context where the room is being managed. For example:

await session.start(agent=agent, room=ctx.room, room_io=room_io)

This helps ensure that the updated

RoomInputOptions

configuration is properly applied to the session. It allows agents to detect each other and makes the

set_name

method work just as you’d expect. If anything’s unclear or you’d like a bit more help, please don’t hesitate to ask—I’m here for you!

this is causing other errors to pop up. if the participant kind is he issue, i am already giveng that in the session.start function explicitly right?

You're absolutely right that specifying

participant_kinds

in the

session.start

function directly sets which participant kinds the agent will process. This parameter works alongside

RoomInputOptions

and doesn’t override it, but instead determines the participant kinds the agent will auto-subscribe to during the session. If you’ve already included

PARTICIPANT_KIND_AGENT

in the

participant_kinds

list within

session.start

, then the issue might lie elsewhere. Could you please share the specific errors you’re seeing? That way, I can better understand what’s going on and help you find the right solution. Sources: metrics/metrics_tts.py | examples/voice_agents/weather_agent.py

there is no error. the issue is that the agent name is not being set correctly

Thank you so much for the clarification! It sounds like the agent name not being set correctly might be related to the timing of when

ctx.agent.set_name

is called. Here are a few things to check that might help: • Make sure the agent is fully connected to the room before you call

set_name

. • Double-check that the

set_name

method is being called on the right agent instance. • Look out for any conflicting calls or overrides that could be resetting the name after it’s set. To dig a little deeper, you could try logging the agent’s name right after calling

set_name

. This way, you can see if the name updates locally but isn’t showing up in the room. Please feel free to let me know how it goes or if you’d like me to help you investigate further—I’m here to support you!

I tried printing the name i set for agent like this:

# Start the session
    await session.start(agent=agent, room=ctx.room, room_input_options=RoomInputOptions(participant_kinds=[rtc.ParticipantKind.PARTICIPANT_KIND_AGENT]))

    await ctx.connect()

    await asyncio.sleep(SLEEP_1)

    await ctx.agent.set_name("Interviewer")

    <http://logger.info|logger.info>(f"Interviewer agent started with name {ctx.agent.name}")

It cam eblank.

It looks like the reason

ctx.agent.name

is coming up blank is probably because the agent's name isn’t being set when the initialization happens. In LiveKit, you’ll want to define the agent’s name right when you accept a job request using the

req.accept()

method. Here’s a little example to help clarify:

async def request_fnc(req: JobRequest):
    await req.accept(
        name="Interviewer",  # This sets the agent's name
        identity="identity",
        attributes={"myagent": "rocks"},
    )

Just make sure the

name

parameter is included during the job acceptance step. If it’s missing there, the name will stay blank. Please feel free to reach out if you’d like me to walk you through it or if you have any other questions—I’m here to help! Sources: Anatomy of an Agent | LiveKit Docs | livekit/src/room/participant/local_participant.rs | Runtime/Scripts/Participant.cs | livekit-rtc/livekit/rtc/_proto/room_pb2.pyi | SIP troubleshooting guide | LiveKit Docs

if __name__ == "__main__":
    cli.run_app(WorkerOptions(
        agent_name="interviewer",
        prewarm_fnc=prewarm,
        entrypoint_fnc=entrypoint
    ))

This is how i am running the agent right now. i am providing the agent_name.

Thanks for sharing about the

agent_name

in

WorkerOptions

! Setting this does indeed assign the agent's name in LiveKit. This setup is especially useful for explicit agent dispatch, which means the agent will join rooms only when they’re explicitly dispatched through the

AgentDispatchService

API, a dispatch rule, or via metadata in a participant's access token. If you’re not seeing the agent name as expected, here are a couple of things to double-check: • Make sure the

agent_name

exactly matches the name used in your dispatch rules or API calls. • Check if there might be any conflicting settings or code that could be resetting the name after initialization. Please feel free to reach out if you’d like me to help you troubleshoot further or if the issue continues—I’m here to help! Sources: Dispatching agents | LiveKit Docs | Agent dispatch | LiveKit Docs | livekit-agents/livekit/agents/worker.py | agents/src/worker.ts | Agents telephony integration | LiveKit Docs

tried giving name through request function. That also not working. I am still getting the name to be "simulated_agent_98983"

I don't have the answer you're looking for. You could also try asking your question: • in one of the other Slack channels or • to https://deepwiki.com/livekit/livekit_composite which is trained on all LiveKit source code If you find the answer, please post it here to help others!