Tailgating at the badge door. A plate that should open the gate. A visitor nobody greeted. These are not security incidents, they are unfinished work, and the camera that watched them happen is the cheapest sensor on site. Nurby turns what it sees into an action in the system you already run.
Every workplace stack has the same hole in the middle. The systems are excellent at recording intent and blind to what actually walked through the door.
Access control logs credentials, not bodies. Every tailgate is invisible to the system that is supposed to be controlling the door, which means your access report is confidently wrong.
Unstaffed hours are when the exceptions happen and when there is nobody to notice them. The footage exists, but somebody has to go looking, and nobody ever does.
The recording proves what happened after you already lost the money. Nothing in the stack turns what the lens saw into a fee, a ticket, or an open barrier while it still matters.
Nurby sits between the lens and your operational systems. It decides on your own hardware and writes the result into the software your team already lives in.
A tripwire on the lane counts everyone who passes. Two people on one credential is an event with a clip attached, on the right account, in the right system.
A plate allow-list drives a dry-contact relay. Known vehicles open the barrier. Unknown ones hold and raise the gatehouse. No keypad and no radio.
The API Call action posts authenticated HTTP with templated bodies, so a rule can add the fee, open the ticket, or stamp the visit wherever your site keeps that record.
Each one is a sentence first. Open any row to see the chain it fires and the primitives it is built from.
Every rule above is built from triggers and actions that ship in Nurby today. Nothing here is a roadmap item.
Nurby's API Call action speaks plain authenticated HTTP with templated URLs, headers, and bodies, so a rule can write into whatever your site actually runs. No certified-partner list to wait for, and no vendor lock-in on our side.
Keyless entry, door controllers, and credential readers. Nurby posts the body count that the badge event alone cannot give you.
Every site is odd in its own way. The lighting, the angle, the one forklift that always parks in shot. A camera product that cannot absorb your corrections makes you tune it forever. Here is where Nurby is on closing that loop, honestly labelled.
A rule can run a vision model as a second opinion before it fires, so the borderline calls get looked at twice instead of paging the shift lead.
Every fired event carries a right/wrong control. Correcting one is the whole interaction. No labelling tool, no dataset, no ML team.
Nurby reads back your corrections and proposes the change: raise this rule's confidence to 0.86 and 28 of the 31 you dismissed would have stayed quiet. You approve it, it does not apply itself.
Your corrections become examples attached to that rule's prompt, so the model learns what a mis-slotted pallet looks like in your aisle, not in a benchmark.
Start with a single camera and a single rule. Prove it on the door that costs you the most, then widen.
Anything that speaks RTSP or ONVIF. No rip-and-replace, no proprietary NVR, no per-camera licence. The hardware on your walls is already enough.
Write what should happen the way you would say it to a new hire. Nurby builds the trigger, the zone, and the action chain, and shows you what it built before it runs.
Point the action at your access, visitor, or billing system. Nurby handles auth, templating, and retries, and logs every fire with the clip that caused it.
Self-hosted, open source, and yours to modify. No appliance to buy, no truck roll to schedule, and your footage never leaves your hardware.