← select a server
Server
Overview
Config
AI Features
Hold Queue
Logs
Mutes
NL Rules
Premium
—
Total Actions
—
Spam Blocked
—
Scams Removed
—
Holds Active
—
Mutes Active
—
Premium Users
Recent Activity
Raid Protection
Log Channel
Modules
Webhook Spam Filter
Deletes messages from webhooks matching spam patterns
Scam Image Reporter
@mention bot with scam image to remove + reward reporter
⚡ AI features require Ultimate tier and a valid
OPENAI_API_KEY in your .env. Set tier to ultimate in Config to unlock.🔍 Scam Image Detection Vision AI
dHash fast-filter first. If uncertain, sends to GPT-4o vision. Detects rotated, cropped, recolored fakes — MrBeast, crypto, nitro.
AI Scam Image Detection
Upgrade from dHash-only to full visual understanding
👤 Fake Account Scoring AI
Score 0–100 on every join. <30 = auto-kick. 30–60 = flagged to log channel. >60 = clean. Uses username entropy, account age, avatar type, join velocity.
AI Fake Account Scoring
Score every join, auto-quarantine high-risk accounts
🧠 Toxicity Detection Context-Aware AI
"idiot-proof guide" = 0. "you're an idiot" = 91. Understands context. Messages above threshold go to hold queue — never auto-banned.
AI Toxicity Detection
Route toxic messages to hold queue for mod review
Messages flagged by AI toxicity detection sit here. Approve to re-post the message, Dismiss to drop it silently. Auto-released after the hold window if no action.
| Action | Target | By | Reason | When |
|---|
No actions logged yet.
| User ID | Muted By | Reason | Expires | Action |
|---|
No active mutes.
⚡ Natural language rules require Ultimate tier. Write plain English — the AI parses it into a rule. Examples: "if someone mentions a crypto wallet address, flag to mods" · "if a new account under 7 days old sends a link, hold for review"
Add Rule
Grant Premium
| User ID | Granted | Expires | Status | Reason |
|---|
No premium grants yet.