Safety

OpenAI-compatible Moderations API.

Create Moderation

Classifies if text inputs are potentially harmful. OpenAI-compatible endpoint.

Run Shield

Run a safety shield on messages to check for policy violations.