Skip to content
Camnitive

Detect

Rule playground

How does a sentence become a camera rule?

Write a safety or compliance policy in plain English and see the deterministic rule it becomes. It runs in your browser.

Your policy

Write it the way you would brief a supervisor: who, where, how close, when, and who should be told.

Compiles as you type.132 of 280

Or start from an example

Compiled rule

person_proximity_forklift_no_reflective_vest

Same sentence, same rule, every time. High severity

rule: person_proximity_forklift_no_reflective_vest

when:

  • class_is: person
  • proximity_lt: 2.0 m of forklift, state: moving
  • ppe_missing: reflective_vest
  • in_roi: zone:two_metres_of_a_moving

then:

  • trigger_after: 2 events / shift
  • severity: high
  • notify: shift_supervisor
  • escalate_to: shift_supervisor

evidence:

  • clip: annotated, −10 s / +10 s
  • explanation: AI-written, in policy language
  • seal: timestamped, tamper-evident

What an event looks like

Camera 14 · person_proximity_forklift_no_ref

A person came within 2 m of a forklift, without reflective vest, in two metres of a moving. Rule fired deterministically; annotated clip sealed and routed.

High severityEvidence sealed

Run this rule on your cameras

We take this exact rule into a fixed-scope pilot on the cameras you already own.

A demonstration of the authoring experience: a keyword parser, not the production compiler. In FloFrame, rules are refined against your zones and cameras.

Compiled rule

person_proximity_forklift_no_reflective_vest

Severity

high

Results

From a sentence to sealed evidence

  1. Step 1

    Describe the policy

    Write the rule the way you would brief a supervisor: subjects, zones, distances, times, and who should be told.

  2. Step 2

    See it compile

    The sentence becomes a deterministic rule: explicit conditions, severity, routing and evidence. The same input always makes the same call.

  3. Step 3

    Run it on your cameras

    In a deployment, the rule runs on FloFrame against the cameras you already own. No model training, no training data to collect.

See it on your own cameras.

A demo on your use cases, or a pilot with success criteria agreed up front.