Skip to main content

Content moderation overview

The Chat SDK supports content moderation for IM messages.

  • The IM service includes a built-in profanity filter for basic text protection (default limit: 50 words). This is not a substitute for professional content moderation.
  • For custom moderation, use pre-messaging webhooks to integrate third-party moderation solutions.

Profanity filter​

tip

The client SDK does not provide management APIs for the profanity filter. It only provides callback APIs to notify the sender when a message is blocked. See Blocked message callback.

The profanity filter identifies and filters sensitive words in text messages. Matched messages can be blocked (not delivered) or have sensitive words replaced.

Supported languages: Chinese, English, Japanese, German, Russian, Korean, Arabic.

Default behavior​

  • Default limit: 50 sensitive words.
  • Only applies to client SDK messages by default.
  • Only applies to the built-in text message type (RC:TxtMsg).

Configuration​

Enable Moderate messages sent by server API in the console.

Pre-messaging webhooks​

For custom or third-party moderation, subscribe to the *:pre_messaging webhook events in the Console under Chat > Webhooks. See pre-messaging webhooks for details.

Last updated on