Profanity filter
The profanity filter is a basic protection mechanism that detects sensitive or inappropriate words in text message content. When a message matches a configured word, Nexconn can block the message so it is not delivered, or replace the matched word before delivery.
The client SDK does not provide management APIs for the profanity filter. The SDK only provides callbacks that let the sender know when a message is blocked. See the sensitive content interception callback page for your SDK.
Supported languages for profanity filtering: Chinese, English, Japanese, German, Russian, Korean, and Arabic.
Manage the profanity word list for each App Key environment through the Server API:
| Action | Server API |
|---|---|
| Add profanity words with replacement content | Batch add profanity words |
| Remove a profanity word | Remove a profanity word |
| Remove multiple profanity words | Batch remove profanity words |
| List profanity words and replacement content | List profanity words |
Default behavior
- The default limit is 50 profanity words.
- The filter applies only to messages sent from client SDKs by default.
- The filter applies only to the built-in text message type (
RC:TxtMsg) by default. - Direct, group, open, and community channels are supported.
- In community channels, edited text message content is also subject to profanity filtering.
- The sender is not notified by default when a message is blocked. Enable the SDK-side sensitive content interception callback if your app needs to update the sender UI.
The profanity filter is built in and does not require activation. By default, moderation features evaluate messages sent from client SDKs. To also moderate messages sent through the Server API, enable Moderate messages sent by server API in the Console under Chat > Chat settings > Security & Moderation. This setting controls whether Server API messages are evaluated by configured and applicable moderation features, including the profanity filter and automatic moderation.