Your agent.
On FaceTime.

Connect your own agent to FaceTime audio and video with Chert’s API, CLI, and LiveKit integration.

Watch the demoChert - Vapi for FaceTime: AI video agents in a few lines | Product Hunt
FaceTime call showing a person and an AI avatar

Some problems
are easier to show.

On a line with caller-camera capture enabled, your agent can use video as input. Confirm that capability separately from sending agent video.

Illustration of a home router with a blue cable in the second Ethernet port from the left
Voice only

“Uh, which cable?”

FaceTime

“The blue cable, second from the left.”

Set up your agent.

  1. Build your agent

  2. Connect your LiveKit room

  3. Test your assigned line

Work that needs eyes.

  • Remote support
  • Field service
  • Telehealth intake
  • Guided onboarding
  • Visual inspections
  • Customer success

FAQ

How does the FaceTime API work?

Chert bridges an assigned FaceTime line to your LiveKit room. You own the agent, room, and conversation. The CLI supports named LiveKit agents; the API supports a backend-controlled integration.

Can I send and receive calls?

Inbound and outbound calling require an enabled, ready line. Caller access and limits depend on your plan. See pricing and the outbound reference.

Does video include caller vision?

Agent video and caller-camera capture are separate capabilities. Confirm both directions for your assigned line and test your agent’s vision integration before relying on it.

Is Chert a replacement for LiveKit?

No. In the Bring Your Own Room integration, Chert connects the FaceTime endpoint to a LiveKit room you control. Read how they work together.

FaceTime API

Build an agent people
can actually show things to.

Start with the setup guide, then verify a real audio or video call on your assigned line.