Moderation Flags

Below is a reference table of the moderation categories used by the AI chat moderator. These categories are used to classify messages detected in server chat and help determine which moderation rules should be applied.

Category	Description
harassment	Content that insults, demeans, or targets a person or group with abusive or harassing language.
harassment/threatening	Harassment that also includes threats of violence or serious harm toward the target.
hate	Content that promotes hatred or discrimination toward protected groups (e.g., race, religion, gender, nationality, sexual orientation, disability).
hate/threatening	Hateful content that also includes threats or calls for violence against a protected group.
illicit	Content that provides instructions or advice for committing illegal activities (e.g., “how to shoplift”).
illicit/violent	Instructions or advice for illegal activities that involve violence or weapons.
self-harm	Content referencing or depicting self-harm behaviors such as suicide, cutting, or eating disorders.
self-harm/intent	Content where a person expresses intent to harm themselves.
self-harm/instructions	Content that provides instructions or encouragement for self-harm or suicide.
sexual	Explicit sexual content or content intended to arouse sexual interest (excluding educational contexts).
sexual/minors	Sexual content involving individuals under the age of 18.
violence	Content depicting or describing physical violence, injury, or death.
violence/graphic	Graphic or highly explicit depictions of violence or injury.