Anti-nuke

What Is a Discord Nuke and How Do You Prevent One?

A Discord nuke is an attack, often scripted, by an account with permissions in your server. It deletes channels and roles, bans members, and floods the rest with webhooks, often within a minute or two. Prevent it with an anti-nuke bot that limits what one account can do in a short window, plus 2FA, tight permissions, and a restore plan.

By Sina, Founder, Security BotUpdated (published )8 min read

Discord nuke meaning

Raids come from outside: accounts join and spam. Nukes come from inside: an account that has Administrator, Manage Channels, Manage Roles, or Ban Members uses those permissions destructively. Because the account is trusted, Discord lets it act.

Required 2FA for moderators and the 2FA or password prompt on bulk bans from the Members page add friction. But nothing in Discord caps how many actions one account can take, and the audit log records the damage after the fact.

A typical nuke script, often called a nuke bot, runs a loop: delete all channels, delete all roles, ban every member it can reach, create dozens of channels with an insulting name, and spam them through webhooks. On a large server the whole sequence can take less time than it takes a moderator to notice.

How nukes happen

CauseHow it starts
Phished staff accountA moderator clicks a fake Nitro, giveaway, or Steam link, or approves a QR-code login, and the attacker takes over the session
Malicious bot applicationSomeone with Manage Server adds a bot that turns out to be a nuke bot
InsiderA staff member leaves angry or was planted by a raid group
Over-permissioned memberA regular member was given Manage Channels for one task and kept it

Note the pattern: every cause is an account you granted permissions to. A nuke is a permissions problem before it is a bot problem.

Warning signs before a nuke

  • A staff member reports their account acting strangely, or their token was in a leaked database.
  • A new bot was added that nobody remembers approving.
  • A member with permissions suddenly creates a test channel or role at an odd hour.
  • Webhooks appear in channels that never had them.
  • The server was recently mentioned in a raid or nuke community.

How to prevent a Discord nuke

Prevention has two halves. First, make the trusted accounts harder to steal and give them less to steal. Second, put a limit on how much damage any account, stolen or not, can do in a short window.

  1. Require 2FA for moderator actions. Server Settings, Safety Setup. Staff without 2FA cannot ban, kick, or manage channels until they enable it. A stolen password stops being enough, though a hijacked session on an account that already has 2FA still gets through.
  2. Apply least privilege. Nobody but the owner needs Administrator. Moderators can ban and lock through a bot with limits instead of holding Ban Members and Manage Channels natively.
  3. Limit who can add bots. Manage Server is the permission that adds applications. Keep it to the owner and one trusted admin, and keep Anti Bot Add on so an unexpected bot addition is punished and the bot is kicked.
  4. Run an anti-nuke bot with per-action limits. Security Bot counts bans, kicks, and channel and role creates and deletes per user, and punishes the account that reaches a limit. Counters reset every 10 minutes. Free instant triggers cover bot adds, prunes, admin role assignment, and dangerous permission edits. Webhook creation is a Premium trigger.
  5. Keep the bot's role at the top. Anti-nuke can only act on accounts below it. If a compromised admin is above the bot, the bot is a spectator.
  6. Enable Beast Mode. A server-wide threat score catches attacks spread over several accounts or paced to stay under each per-user limit.
  7. Enable Recovery in advance. On Premium, Recovery takes automatic backups of the server layout (roles, channels, permissions, emojis, and who had which role). It saves every role, channel, category, and emoji the moment it is deleted. After a nuke you restore what the attacker deleted, or the last backup from before the attack, and members get their roles back. Anything from before you turned it on is not saved.
  1. Require 2FA for moderator actions

    Server Settings, Safety Setup. Staff without two-factor authentication lose moderation permissions until they enable it. That stops attackers who only have a password. It does not stop a hijacked session, such as a stolen token or an approved QR-code login.

  2. Cut Administrator down to the owner

    Audit every role with Administrator or Manage Roles, Manage Channels, Ban Members. Give staff moderation through a bot instead of native permissions.

  3. Invite Security Bot and put its role on top

    Anti-nuke only acts on accounts below the bot. Drag its role above every staff role.

  4. Set anti-nuke limits and punishments

    Seven free triggers are on by default. Turn on prune, admin role assignment, and dangerous permission edits as well. A ban limit of 2 to 3 and a channel delete limit of 2 suit most teams. Use clear roles as the punishment for staff mistakes and ban for channel and role deletes.

    Command: /setlimit action:Banning Members limit:3

  5. Whitelist the admin role per action

    Exempt your admin role only for the actions it really performs in bulk, one category at a time.

    Command: /whitelist add_role role:@Admin category:Roles: Channel-Delete

  6. Turn on Beast Mode

    Give points to the actions you care about and set a server-wide score limit, so distributed or slow attacks are caught even when no single account reaches a limit.

  7. Enable Recovery

    On Premium, turn Recovery on with the switch on the Recovery page or the command. From then on it backs up the server layout every 24 hours by default and saves every deleted role, channel, category, and emoji. You can then put back what an attacker deleted or changed.

    Command: /recovery enable

What an anti-nuke bot actually does

An anti-nuke bot reads the same events the audit log records, as they happen, and keeps a per-user counter for each dangerous action. When a counter reaches the limit you set, the bot applies a punishment to that user: kick, ban, or clear roles.

Security Bot defaults to kick for most triggers. Clear roles is a sensible choice for staff accounts, because it stops the attack and is reversible if a real staff member triggered it by accident.

Two details matter beyond the toggle. First, the bot's own moderation commands should not become a way around the limits. In Security Bot, while the Anti Ban and Anti Kick triggers are on, /ban, /softban, /banid, and /kick are capped at your ban or kick limit per moderator per 10 minutes. Further uses are refused until the counter resets.

Second, the bot needs instant triggers for actions nukes often start with: adding a bot and assigning an admin role (free), and creating a webhook (Premium).

Anti-nuke settings by server size

Seven of Security Bot's ten free triggers are on by default: ban, kick, channel create and delete, role create and delete, and bot add.

The default limits are 5 bans, 5 kicks, 3 channel or role deletes, and 8 channel or role creates per user. Counters reset every 10 minutes, and kick is the punishment for most triggers. Tighten them to match how your team actually works. These are suggested values, as in the setup guide:

ServerBan and kick limitsPunishmentAlso turn on
Under 1,000 membersBan and kick limit 2Clear rolesOne-click or image captcha verification
1,000 to 20,000 membersBan limit 3, kick limit 3Clear rolesAnti-raid with lockdown (Premium)
20,000+ members or public communitiesBan limit 3Ban on channel and role deleteAdaptive Verification on Strict and Beast Mode

Whichever row you pick, turn on the instant triggers too. They have no limit to tune. Bot add is on by default, while prune, admin role assignment, and dangerous permission edits are off until you enable them. Webhook creation is a Premium trigger.

If a trigger fires on one role or bot during routine work, add a per-action whitelist entry for that role or bot instead of raising the limit for everyone. Raise a limit only when the whole team's normal volume is above it.

If a nuke is happening right now

  1. Remove the attacker's permissions. As owner, strip their roles or ban them. If the attacker is a bot, kick the bot.
  2. Change nothing else yet. Do not start recreating channels while the script may still be running.
  3. Check the audit log for the account responsible and for any bots or webhooks it created. Delete those.
  4. Restore with Recovery if it was on before the attack. On the Recovery page, restore what the attacker deleted from Recently deleted, or restore the last backup taken before the attack. Check the preview first.
  5. Unban members who were banned during the attack. Filter the audit log by the attacker and the Ban Member action to get the list.
  6. Require 2FA, audit permissions, and set limits before the next one. Read how to recover a Discord server after a nuke for the full checklist.

Nuking is against Discord's rules

Nuking a server violates Discord's Terms of Service and Community Guidelines. Report the responsible account through Discord's report form with audit log evidence. Discord does not restore deleted channels, which is why a recovery plan is on you.

Do this with Security Bot

Discord Anti Nuke Bot: Stop nukes with per-user action limits, automatic punishments, owner alerts, and a security log.

Security Bot is a Discord anti nuke bot that watches bans, kicks, channel and role changes, bot adds, prunes, and dangerous permission edits. When an account hits a limit, it is kicked, banned, or stripped of its roles. The server owner gets a DM.

Frequently asked questions

What does nuke mean in Discord?

A destructive attack by an account with permissions that deletes channels and roles and mass-bans members, usually with a script, in seconds.

What is anti-nuke in Discord?

A bot feature that counts dangerous actions per user and punishes any account that reaches a limit, stopping a nuke after the first few actions.

Can Discord restore a nuked server?

No. Discord does not restore deleted channels or roles. A Server Template made earlier can rebuild the structure in a new server. Security Bot's Recovery, if it was on before the attack, restores deleted channels, categories, roles, and emojis from its backups and its recently deleted list.

How do nuke bots get into a server?

Someone with Manage Server adds them, often after being told the bot does something useful. Limit Manage Server and keep the Anti Bot Add trigger on.

Does 2FA prevent nukes?

Requiring 2FA for moderator actions stops attackers who only have a password. It does not stop a hijacked session, such as a stolen token or an approved QR-code login. It also does not stop an insider or a malicious bot, so pair it with anti-nuke limits.

Will anti-nuke punish my moderators for normal work?

Rarely. Limits are per action and counters reset every 10 minutes, so routine moderation stays under them. If your admins delete channels in bulk, whitelist the admin role for that action rather than raising the limit. Raise a limit only when the whole team's normal work reaches it.

How this guide was made: drafted with AI assistance, then checked against Security Bot's source code and Discord's official help articles. Spotted something out of date? Tell us in the support server.