Agentic Presence

Your agent can see who's in the room.

Agent Eyes uses your webcam + on-device face recognition to identify who's in front of the screen. No cloud upload — embeddings stay on your machine. Routes greetings, intel, and conversation history to the right person.

Choose Your Plan Back to overview

On-device face recognition that identifies who's in front of your camera and routes context to them.

What it does

Agent Eyes at a glance.

On-device only

ONNX face-embedding model runs in the home server. Embeddings never leave the box. No cloud face-recognition API ever called.

Per-participant tabs

When a recognized face appears, the conversation engine auto-spawns a participant tab with their history, intel, and shared context.

Family Brain wiring

Recognized family members surface their portion of Family Brain in the right rail — recent posts, photos, gift history, upcoming events.

Pluggable

Bundled npm model for the basics; swap in heavier models (Insightface, ArcFace) when you've got the GPU for it.

Add Agent Eyes to your agent.

Every VP3 plan unlocks the full c:// drive catalog. Pick a plan, install the home server, flip the app on.

Choose Your Plan