Skip to content

Best Practices

timeout187 edited this page Jul 9, 2026 · 1 revision

Best Practices

Recommended roles, hierarchy, whitelist strategy, and tuning for a well-defended AntiLink server.

View this page in the localized AntiLink documentation

AntiLink protects your server out of the box, but a few deliberate choices make it noticeably stronger and reduce false positives. Here is what we recommend.

Get the role hierarchy right

This is the highest-impact setup step. Discord only lets a bot act on members and roles below its own.

Warning

Drag the AntiLink role high in Server Settings -> Roles: above your regular members, and above your mute / quarantine role. If the AntiLink role sits too low, quarantines, mutes, and bans will silently fail.

  • Create a dedicated mute / quarantine role that denies Send Messages, and place it below AntiLink but above normal members.
  • Keep AntiLink's role below your most senior admins if you prefer, but above everyone it needs to moderate.

Set a log channel

Pick a private moderation log channel (staff-only) so every removal, action, and config change is visible in one place. Pair it with the dashboard Logs & Activity view for filtering and history.

Keep your whitelist tight

A smaller whitelist is a safer whitelist. Whitelist only what you actually trust:

  • Your own bots' link domains and any command shorteners they use.
  • Trusted partner invites and community domains.
  • Staff roles that need to post links freely.

Avoid whitelisting broad, public channels unless necessary - a whitelisted channel is an unprotected channel. See the Whitelist guide.

Choose the right enforcement mode

Community type Suggested mode
Most servers Delete - clean and quiet
Strict / high-risk Quarantine - holds offenders for review
Lenient / friendly Warn - escalates only repeat offenders

Set this on the dashboard Protection tab or with /config. See How protection works for how each mode behaves.

Tune Member Defense carefully

If you run Member Defense (AntiLink Premium):

  • Start with the conservative default thresholds and tighten only if you see real raids slipping through.
  • Use a quarantine action rather than ban while you calibrate, so mistakes are reversible.
  • Turn on Emergency Lockdown auto-trigger so a detected raid tightens invite and external-link blocking server-wide automatically.

Place Honeypot channels well

A Honeypot trap catches bots and raiders that post in a channel no real member should:

  • Name it obviously, like #do-not-post-here, and keep it visible near the top.
  • Start with the delete + log action, then move to quarantine once you trust it.
  • Never set it to ban as a first step - the default is deliberately safe.

Always test with a normal account

Because AntiLink exempts owners, admins, and moderators, testing with your own account will show nothing happening. Keep a non-admin test account (or ask a regular member) to confirm changes really work.

Watch your Security Score

The dashboard Security Center gives your server a Security Score and a Permission Health check. Aim for a high score and clear any flagged missing permissions - those flags are the fastest way to spot a misconfiguration before it matters.

Clone this wiki locally