Small decisions. Interesting possibilities.Submit a project / videoSubmit

How it uses Jev

Judges proposed tool calls before execution and tool results before the agent reads them; policy thresholds and decision logs stay outside the prompt.

Why it's interesting

The project exposes missed detections and starts in shadow mode, making its fast safety layer measurable and reviewable instead of silently overriding the agent.

How often does Jev run?

This project calls Jev on demand, in response to user input or events.