Privacy Policy
Effective Date: November 14, 2026
Argus ("the Bot") provides automated moderation and security services for Discord servers, plus an optional web dashboard. This policy explains what data we collect, how we use it, how long we keep it, and your choices.
1. Data We Collect
To function, Argus stores the following in its database:
- Configuration Data: Guild IDs, channel IDs, and role IDs, plus the filter rules, custom word lists, allowed-word lists, and feature settings you configure.
- Moderation Records & Evidence: For each moderation action (warn, timeout, kick, ban, message deletion) we record the involved user IDs, the responsible moderator (or the bot, for automated actions), the reason, the filter that matched, timestamps, and the actions taken. A case may also retain a snapshot of the triggering message and attachment URLs so moderators can review evidence after the original message is deleted. This powers warning thresholds, cases, appeals, and the Analytics dashboard. The username at the time of the action is stored so dashboards render without re-querying Discord.
- Moderator Queue & Reports: If a server uses reports, the join gate, or AI-filter dry run, we store the reported or flagged member IDs, reporter and reviewer IDs, review outcomes, and the message snapshot, report statement, score, or gate evidence needed for moderators to review it.
- Scheduled Actions & Restrictions: We store target and creator IDs, role IDs, reasons, due dates, execution status, retry errors and linked case numbers for temporary bans, verification deadlines and scheduled role changes. Restrictive role records may survive a member leaving so the restriction can be restored if they rejoin before it expires.
- Configuration Audit: We record which administrator changed a setting, when, and which fields changed. Configuration values may be included as before/after values in this audit trail.
- Moderator Notes: Free-text notes your staff voluntarily attach to a user via
/note.
- Ban Appeals: If appeals are enabled, the appeal statement a banned user submits, along with their user ID, the ban reason, and the moderator's review decision.
- Server Backups: If you create a backup, we store its creator ID and username, server owner ID, roles, channels, channel topics and permission overwrites, and server settings. Backups can include copies of server icons, banners and splash images; optional sections include emoji images, the ban list and reasons, and the member list with user IDs, nicknames and role assignments. Backups exist solely to restore your own server.
- Web Dashboard Accounts: If you sign in to the dashboard, we use Discord OAuth2 (
identify and guilds scopes). Your Discord profile (including user ID, username and avatar), server list and OAuth access token are kept in a server-side MongoDB session so you stay logged in. The access token is not a browser cookie.
- Operational & Diagnostic Data: Data about Argus itself rather than about you — which of our servers is currently running the bot, its uptime and memory use, and rotating server-side log files recording startups, shutdowns, errors and gateway disconnections. These logs can incidentally contain identifiers (such as a guild ID) that appear in an error message; they are used to diagnose outages and crashes. Message bodies are not intentionally logged, but error details can incidentally include personal data.
Note on Message Content:
Most message content is processed in memory and discarded. When a message becomes case evidence, a member report, or an AI dry-run queue item, Argus stores a limited text snapshot and attachment URLs so server moderators can review it later. Scanned message images are not stored; backup images are handled separately as described above (see Section 4).
2. How We Use Data
We use this data only to operate Argus: enforcing the moderation rules you configure, counting warnings against your thresholds, rendering your dashboard and analytics, processing ban appeals, restoring backups you request, and keeping the service online and diagnosable. We never sell your data or use it for advertising, and we do not use your data to train machine-learning models.
3. Data Retention
- Analytics records auto-expire 180 days after the action.
- Moderator queue items auto-expire after 90 days, and configuration audit records after 365 days.
- Cases and their evidence, warnings, moderator notes, ban appeals, and configuration persist until you delete them, apply a shorter server retention setting where available, or remove the bot. If you enable warning decay, expired warnings stop counting toward your punishment thresholds but the records themselves are not deleted — they remain visible in
/history until removed.
- Scheduled actions remain while pending or running. Completed, cancelled and failed actions auto-expire 14 days after their completion timestamp. Persisted restrictions are removed when cleared or when their configured expiry is reached; restrictions without an expiry remain until cleared or the server's data is purged.
- Backups are kept until you delete them or they are pruned beyond the maximum you set (default 10).
- When Argus is removed from a server, that server's database records are deleted after a 30-day grace period unless the bot is re-added first. Deletion runs during periodic sweeps, so it is not immediate at the deadline. Database expiry also runs asynchronously. Guild deletion does not erase separate dashboard sessions, rotating host logs, messages already posted in Discord, or copies of exports downloaded by server staff.
- Dashboard sessions use a browser session cookie. Server-side session records expire after 14 days of inactivity under the current session-store settings; closing the browser does not immediately erase those records. Logging out removes the authenticated profile and access token from that session.
- Node uptime history (used to draw the 90-day bars on the status page) expires automatically after 120 days.
- Diagnostic log files rotate on a fixed size limit — a small number of recent files is kept on the host and older files are overwritten as new ones are written.
- Malicious-link results are cached in memory and reused for up to one hour before a fresh check is needed; the cache is never persisted.
4. Message Content & Media
Argus scans messages and image attachments in real time to detect profanity, NSFW material, hate speech, and hostile behaviour. Enabled identity filters also examine usernames and nicknames. Join-gate and optional trust features use account age, server join time, roles, avatar presence and patterns across recent joins to assess members; resulting scores or gate evidence may be retained with moderation records or queue items. File filtering checks attachment names and extensions rather than inspecting file contents for malware.
- Text checked by the built-in filters is processed in memory against local word lists and pattern matchers. It is discarded unless it is retained as case evidence, a member report, or an AI dry-run queue item. The separate AI Server Rules feature is described below.
- Message images checked by the built-in NSFW filter are downloaded from their attachment URLs and classified by a machine-learning model running locally on our servers, without retaining the image bytes. If server staff separately enable AI Server Rules with image scanning, Discord image attachment URLs are sent to an external AI decision service for analysis. Attachment URLs may remain in moderation evidence. Server backup images are stored separately as described in Section 1.
- The Context-Aware (AI) filter — which scores how hostile a message is, and in which of 28 languages — is likewise entirely local. It performs no network calls, downloads no model, and sends nothing to any AI provider. The scoring engine is deterministic; when dry run is enabled, its result and a limited message snapshot are stored in the moderator queue.
- AI Server Rules is a separate, opt-in external AI feature. It sends recent channel text, author IDs and names, reply references, timestamps, configured rules and optional Discord image attachment URLs to an external AI decision service. The default context is up to 30 messages, configurable up to 100, and can include exempt members as context. Staff-requested rule extraction sends the latest 100 rules-channel posts (including embed text) to an external rule-extraction service. Rules are stored in guild settings; preview and enforced decisions retain limited evidence in moderation cases. Pending messages are held in memory while a batch is processed and discarded afterwards.
- To catch abuse split across several messages, Argus may hold a user's recent messages in memory for a short buffer window (about 15 seconds, capped at 10 messages). The buffer is discarded when the window elapses and is never written to disk.
- If you enable the Message Audit Log, edited or deleted message content is posted to a log channel inside your own server (only when the message is still in Discord's cache). It is not written to our database.
- When a filter opens a case, a limited snapshot of the triggering message and attachment URLs may be retained as evidence as described in Section 1.
5. Automated Decision-Making
Argus is an automation tool: deletions, warnings, timeouts, kicks and bans can be applied automatically, without a human reviewing them first, according to the rules you configure. Server staff choose which filters are on, how sensitive they are, and what each one does; they can review recorded actions through cases, the moderator queue, configured log channels and analytics. Some actions can be reversed, but deleted messages cannot be restored, and permissions or API errors may prevent a reversal. Users who believe an action was wrong can raise it with their server's staff, or — where the server has enabled ban appeals — submit an appeal, which a human moderator reviews.
6. Third-Party Services
- Discord: Argus operates on Discord; your use is also governed by Discord's Privacy Policy.
- Google Safe Browsing: When the link filter and an API key are configured, URLs found in messages are sent to Google's Safe Browsing API to check against its database of unsafe sites. Argus does not separately send the author's identity or surrounding message, but a URL itself can contain personal data or tokens in its path or query string.
- Top.gg: When configured, we periodically send Top.gg the bot ID and total server count for listing statistics. No individual user or server identifiers are included.
- Website resources: The public website and dashboard load fonts from Google Fonts. The dashboard also loads interface libraries from unpkg and jsDelivr, and Discord-hosted images may be requested from Discord's CDN. These providers receive your IP address and request metadata when your browser fetches their resources.
An external AI decision service receives configured rules, message context and optional images when AI Server Rules is enabled. An external rule-extraction service receives selected rules-channel text when staff request extraction. The built-in local filters do not send message content to these AI providers. External AI processing occurs only through the separately configured AI Server Rules feature. URLs from messages can be disclosed to Google Safe Browsing as described above. Records and log messages delivered inside Discord are processed by Discord and can be viewed by people who have access to the relevant channels. Server staff control those channel permissions.
7. Cookies
The web dashboard sets a single essential session cookie to keep you signed in. It is Secure, HttpOnly, and SameSite=Lax, and is used only for authentication — never tracking or advertising.
8. Data Storage and Security
Persistent application records and dashboard sessions are stored in MongoDB; diagnostic logs are stored separately on the host. The dashboard uses HTTPS and secure session cookies, write actions are protected against cross-site request forgery and rate-limited, and settings writes are restricted to an explicit list of fields. Database transport security depends on the configured MongoDB connection. Access to stored data is limited to systems and operators who need it to run and support Argus.
Argus may run on more than one server for redundancy (an active node and a standby that takes over if the active one fails). All nodes read from the same database, so your data is not duplicated into separate stores; only one node drives your server's moderation at any moment. Diagnostic log files are stored on the host running that node.
No system is perfectly secure, and we cannot guarantee absolute security. Argus limits persisted message content to moderation evidence and review workflows and applies the retention controls described above.
9. Your Rights and Choices
| Entity |
Rights & Actions |
| Server Owners |
Removing the bot starts the 30-day grace period before guild data is purged; request a manual wipe for earlier deletion. Set shorter retention periods for analytics, queue items, cases and warnings, and control optional features such as appeals, backups, message logs and AI filtering. |
| Individual Users |
Use /dataexport to privately export your own retained records in the current server. Moderators with Moderate Members permission may export another member's records. Exports include subject references in backups and redact activity involving other members or confidential settings. Ask server staff to review or delete records about you, or contact us for access or deletion requests. |
| Dashboard Users |
Log out to end your session, and revoke Argus's access anytime from Discord's Authorized Apps settings. |
Depending on where you live, you may have additional rights over your personal data — access, correction, deletion, and objection among them. Contact us (Section 11) to exercise them; we aim to respond within 30 days. Because Argus identifies people only by their Discord user ID, we may ask you to verify that ID through Discord before acting on a request.
10. Children's Privacy
Argus is not directed to anyone under 13 (or the minimum age of digital consent in your country). We do not knowingly collect data from children below that age. If you believe a child has provided us data, contact us and we will delete it.
11. Changes and Contact
We may update this policy; changes are reflected by a new effective date at the top of this page. Review it periodically to understand current data practices. This notice does not replace any consent required by applicable law. For privacy inquiries or data-deletion requests, contact us at contact@superbot.click, or reach the bot owner on Discord (1121394601333370981).