All guides

Protecting a Discord server from raids and fake accounts

A raid means having a large number of accounts join a server in a short time, to flood its channels or mass-mention its members.

The right defence is not detecting the raid as it starts, but making sure a freshly arrived account cannot write anything until a human has let it in.

Last updated:

The layer Discord provides

In the server settings, the verification level sets conditions on writing: a verified email address, an account older than five minutes, or membership of the server for more than ten minutes. The highest level requires a verified phone number.

This layer is free, immediate, and stops the least sophisticated waves. It is not enough on its own: aged, verified accounts are resold, and a prepared raid uses them.

Close the server by default

This is the most effective measure, and the most often skipped. Remove from the @everyone role the permission to view channels, except for a single landing channel. An account joining the server then has nothing to flood.

Real access comes from a separate role, granted only after verification. Until that role is given, the newcomer sees a closed door — however many accounts arrived at the same time.

Approve arrivals by hand

With Eric's verification module, each arrival posts a validation request in a staff-only channel. A moderator reacts to accept, and the access role is granted automatically.

The value of this approach is that it rests on no heuristic. Automatic filters — account age, missing avatar, suspicious username pattern — are worked around by preparing accounts in advance. A moderator watching forty identical requests arrive in two minutes is not fooled.

What to prepare before the incident

  • An access role distinct from @everyone, with channel permissions that genuinely depend on it.
  • The bot's role placed above the access role in the hierarchy.
  • A validation channel visible to staff only.
  • Discord's verification level set to at least medium.
  • A written instruction for the team: who raises the verification level, and when.

During a raid

  • Immediately raise the server's verification level to its maximum: this blocks writing from accounts without a verified phone number.
  • Suspend active invites if the raid arrives through a publicly shared link.
  • Ban rather than kick: a kick lets them return with the same invite.
  • Do not delete the messages before collecting the account identifiers, which are needed to report them to Discord.
  • Restore the settings once the wave has passed, or genuine new members stay locked out.

Frequently asked questions

How do I stop fake accounts from joining a Discord server?

You do not stop them joining: Discord does not allow it. You stop them acting, by removing channel access from the @everyone role and granting the access role only after a moderator approves.

Is Discord's verification level enough against a raid?

Not on its own. It stops improvised waves, but aged accounts verified by email and phone are resold. Combine it with a server closed by default and human approval of arrivals.

What should I do immediately during a raid?

Raise the verification level to maximum, suspend publicly shared invites, ban rather than kick, and collect the account identifiers before deleting messages so you can report them to Discord.