Agent tooling

JEV Safety Gateway

A Go reverse proxy that sits between nginx and your LLM backend, judging every request's user input with Jev and blocking harmful traffic transparently.

Bookmark Keep this in your collection.

Leave a noteWhat would you try with this?

Only you can see your notes.

Source screenshot of JEV Safety Gateway — SOURCE SCREENSHOTFull screenshot ↗

What it does

It extracts the input by API path, ships with a Vue 3 admin UI and no-CGO SQLite storage, and deploys via Docker on Linux and Windows.

How you can use it

Before writing any code, list the kinds of messages your AI app handles, such as customer questions or story prompts. Note any sensitive personal details you must not share with third parties. A developer can then set up this filtering program between your web traffic and your AI service.

Your developer will need an access key that connects the gateway to the TypeSafe safety service. Keep in mind that automated checks can make mistakes, and every incoming message gets sent to that external service for review, so it should not be treated as a guarantee of complete safety.