Transparency

What We Collect

• User IDs and Discord Usernames to maintain the blacklist system.
• When a ban occurs, the bot saves the Server ID where the event triggered to help staff investigate raids and take a deeper dive into specific scams.
• We store specific message content or Message ID that triggered a detection (e.g. a scam link).
• If an image is detected as a scam, we store its hash for future detection. Images that are not detected are not stored at all. We never store the original image, only its mathematical hash.
• For users flagged as "Hacked", we temporarily store a list of your roles. This allows us to restore your permissions once your account is secured and the appeal is granted.
• Any notes added by staff members during a ticket or appeal are stored to ensure consistent moderation.

How We Use Your Data

• Data is shared across the servers to block known bot accounts and scammers instantly.
• We track detection counts (e.g. how many scams were blocked today) to monitor performance. These stats do NOT identify individual clean users.
• Your Ban ID v2 XXXX-XXXX-XXXX-XXXX-XXXX-XXXX is used to link your appeal ticket to your specific case in the database.
• If a user is re-flagged after a previous ban, the prior case details are kept in that user's history rather than overwritten.
• Ticket conversations are archived in our database for quality assurance.
• An big part of our database is private and accessible only by authorized administrators.
• Users who have never triggered a security flag do not have personal data stored in our permanent ban databases.
• Our public database is located at TokBot/blob/main/scammers.json, sensitive data is not publicly shared.

AI Content Scanning

Judicium uses automated AI models to review messages and images in servers where protection is enabled. These checks look for:

• Child sexual abuse material (CSAM), which is blocked immediately and reported according to the law.
• Sexually explicit (NSFW) content.
• Graphic violence and gore.
• Toxic or abusive text, and scam or NSFW language.

Scanning happens while the content is being checked. Messages are not saved for later review, and the AI models do not keep a copy of what they look at.

Real-Time Message Processing

• A message is checked in memory only.
• If a threat is found, the user may be banned and the message removed.
• If the message is safe, it is discarded right away.
• Messages are never written to disk as part of normal operation.

What We Don't Collect

• Full message text, unless it was part of a detection.
• Direct messages or private channel content.
• IP addresses or connection data.
• Email addresses or other contact information.
• Full images. Hashes are only kept for images that were detected as a threat.
• Server configuration or channel settings.

How Long We Keep Data

• Cases, server blacklists, scammer records, ban records, exploit attempt records, MSID records, snapshots, user history, and audit logs are kept until they are removed manually. They are not automatically deleted.
• Bot logs, debug logs, console logs, token audit logs, and daily analytics are deleted automatically after 365 days. Only entries older than 365 days are removed. Newer entries remain.
• Rotated bot log backups are deleted automatically when their last write is over 365 days old.
• Message contents are deleted from Discord when a moderation action purges a user's messages. The bot does not keep a separate copy of message contents in its logs.

You can ask for case data to be removed by appealing in our support server.

Data Security

All data is encrypted in transit, and we use standard practices to protect it from unauthorized access. We do not sell or share user data with third parties.

The /rejoin Command

The /rejoin command lets a user who was removed from a server request to come back. It uses Discord's own login (OAuth2) so the user approves each step.

• Only the permission needed to add the user back to the server is requested. Judicium cannot read your messages or DMs through it.
• Each request expires after 5 minutes and can only be started once every 60 seconds.
• Blacklisted users cannot use /rejoin.
• No permanent record of rejoin requests is kept. The access token is used once and then discarded.

Your Options

• You can appeal a ban or hacked-account flag through our support server.
• You can ask for a copy of, or the removal of, data tied to your case through the support server.
• If your appeal is granted, the record of that case is updated to match.