What it does
Its default endpoint is a community-run Jev gateway rather than TypeSafe's own, which the repository describes as the TypeSafe cloud API. With no key it runs a simulated engine.
Agent tooling
Jev Decision Engine
An agent skill and stdlib CLI that turn Python or JavaScript question builders into one System One payload, with six named decision playbooks.
Bookmark: Jev Decision Engine
Keep this in your collection.
Leave a noteWhat would you try with this?
: Jev Decision Engine
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can adapt this project to check text commands for potential risks. The provided code includes a test that examines a request to delete files. The system sends the text to an external AI service. The software blocks actions marked as dangerous, but AI choices can still be wrong.
The external service returns a choice and a score for permanent damage. It also guesses whether a person should approve the action. Working software requires an access key that connects the app to the external service. Without this key, the tool only runs a limited simulation.