Loading...
Start with Balanced Protection unless your server is already under attack.
Open Test Rules and try normal messages, spam, invite links, and suspicious links.
Review Settings before turning on harsher actions like mutes, kicks, or bans.
Check Activity & Evidence after the first day and loosen anything catching normal users.
Run the simulator before strict enforcement by using Test Rules. If a rule punishes your sample normal messages, it is not ready for live heat.
Use scoped exemptions before global bypasses. Let trusted channels breathe without making the whole server softer.
Evidence is redacted by default. You get enough context to moderate without turning the dashboard into a raw message archive.
No infrastructure metrics belong here. AutoMod shows moderation outcomes, not database status, worker health, or other backend guts.
Your live command center: protection status, recent actions, triggered rules, incidents, recommendations, active protections, and quick actions.
Global behavior: mode, evidence privacy, staff bypass, public replies, message deletion, and always-block settings for dangerous scams and raids.
High-level protections for phishing, raid defense, new accounts, invite abuse, attachments, and mass mentions.
Fine-tune each rule's enabled state, base action, threshold, and punishment ladder.
Allow trusted roles, users, channels, or categories to bypass normal rules without disabling protection server-wide.
Review redacted evidence, matched rule names, actions taken, reasons, and timing without exposing full message content.
Apply a sane baseline like Balanced Protection, Strict Security, Raid Defense, or Community Friendly.
Test a message and user context before changing live enforcement. This is your safest way to avoid accidental chaos.
Track or score the event without punishing the user.
Good for first-time or low-risk violations.
Best for spam pressure, repeated violations, and raid cleanup.
Use when the account looks hostile but a ban is too aggressive.
Reserve for critical phishing, raid accounts, and clear malicious behavior.
Use trusted roles, users, channels, and categories when a rule is right most of the time but too strict for one area. Keep critical phishing and raid protections visible so staff bypasses do not become a free pass for dangerous behavior.
Evidence should tell you what rule matched, what action happened, and why. It should not expose full raw conversations by default. Use the Evidence tab to audit behavior and tune thresholds.
Raid Mode is for active pressure: mass mentions, throwaway accounts, and fast-moving spam waves. It requires confirmation because it intentionally tightens enforcement for a short window.
Pick your guild, open AutoMod, apply a preset, test a few messages, and then watch Activity & Evidence after real traffic hits.