SDKs, API and evaluation
The primitives every integration above is built on, plus offline evaluation for your test sets.
The wonderfence-sdk package wraps the Alice evaluate API. Every call returns an action (BLOCK, MASK, DETECT or NO_ACTION), the masked or block text, and the detections that fired, so your code decides what to do next.
Every SDK and gateway integration on this page ends up at the same evaluate endpoint. If your stack is not covered, call it directly from any language with an API key and an application id.
An MLflow scorer that sends each row of an evaluation dataset to Alice and records the verdict as feedback. Compare models or prompt versions on safety the same way you compare them on quality.
A TypeScript client for the Alice evaluate API. One client instance serves many applications because appId is passed per request. Built for Node.js sandboxes such as Cognigy extensions and serverless functions.