Peer-to-peer video with one person at a time. Nothing recorded. Leave, and it’s gone.
Video flows browser-to-browser over WebRTC. We don’t see frames.
Nothing stored on our servers. Only a hash, and only if you report.
Hard gate on entry. Cloudflare CSAM Scanning at the zone.
Permission, pair, peer-to-peer.
Your browser proves permissions are live before the queue accepts you. We mint a two-minute ticket — no recording, no tracking.
The video queue only contains other video users. WebRTC negotiates a direct connection. When direct fails, Cloudflare Realtime relays ephemerally.
Report captures a perceptual hash in your browser — the frame never leaves your device. Close the tab and the room is destroyed.
No. A single 18+ confirmation and a camera/mic permission grant is the full onboarding.
Not by us. The video stream is peer-to-peer or relayed ephemerally; our servers never store frames. If you press Report, only a perceptual hash of the frame is sent — not the frame itself.
The queue only accepts users who already have working camera and mic. This prevents one person with a broken mic from getting paired repeatedly. Your browser proves the permission is live; we issue a two-minute ticket that the queue verifies.
No. Video and text are separate queues. A text user will never be paired with a video user under any condition.
Signaling uses Cloudflare Realtime as a TURN fallback. If direct peer-to-peer fails, your stream is relayed via the nearest Cloudflare point-of-presence with short-lived HMAC credentials.