Content Filter is a bot that uses OpenAI's GPT-3 content filter in order to filter out potentially unsafe or sensitive content. This is the perfect bot to maintain a positive atmosphere and keep a family-friendly server safe!

Usage

The bot will automatically start moderating the server once added. The default settings will remove any messages that it deems are unsafe from members without the "Manage Messages" permission.

You can use the commands below to customize the behavior of the bot.

• /filter get: Displays server's current filter level.
• /filter set: Sets how strict the filter is on the server. Requires "Manage Server" permissions.
• /reveal get: Gets what [Removed] messages can be revealed on the server.
• /reveal set: Sets what [Removed] messages can be revealed on this server.
• /help: Opens the help menu.

Filtering

There are three levels of filtering:

• Strict:
  • Profane language
  • Prejudiced or hateful language
  • Something that could be NSFW
  • Text that portrays certain groups/people in a harmful manner
• Default:
  • A sensitive topic
  • Something political
  • Something religious
  • Talk about a protected class such as race or nationality
• None:
  • Disables the filter on the server

Revealing

Users can reveal the original content of [Removed] messages by right clicking a message, then selecting Apps -> Reveal. The behavior of this command can be modified with the /reveal set command.

If you do not like the [Removed] insertions, you can disable this feature by revoking the bot's "Manage Webhooks" permission.