Verification

How to Stop Bots From Joining Your Discord Server

To stop bots from joining a Discord server, combine a higher verification level with captcha verification so automated accounts cannot post. Use Discord's Raid Protection and, on Premium, anti-raid scoring to detect and punish join waves. Keep Anti Bot Add on to punish anyone ranked below the bot who adds a bot application. Pause invites as the emergency brake.

By Sina, Founder, Security BotUpdated (published )9 min read

Two different kinds of bots

People say bots when they mean two things:

  • Bot accounts (user bots or self-bots) are normal Discord accounts driven by a script. They join through your invite link, post spam or scam links, and mass-mention.
  • Bot applications are real Discord bots that someone with Manage Server adds to the server. A malicious one is a common start of a nuke.

The fixes are different. Bot accounts are stopped at the join step with verification and join protection. Bot applications are stopped by controlling who can add them and punishing anyone who does without permission.

Why bot accounts keep finding your server

  • Your invite link is public: on a website, a bot list, social media, or Server Discovery.
  • Someone in the server leaked an invite into a raid community.
  • Your server has a custom invite link (vanity URL, a Boost Level 3 perk) that is easy to guess.
  • The server was raided before and the old invite is still being passed around.

You cannot make an invite link unfindable and still grow, so the goal is to make joining pointless: the account gets in and cannot post in your channels. It can still DM your members, which Rules Screening, Security Actions > Pause DMs, and Discord's DM-spam Activity Alerts help with.

Step 1: Set the verification level to Medium or High

Open Server Settings > Safety Setup (on a Community server, DM and Spam Protection) and set the verification level. Medium requires the account to be at least five minutes old. High also requires ten minutes in the server before posting.

Accounts created moments before a raid are held back by Medium, and High turns join-and-post into join-and-wait. A verified phone number skips every level requirement, so neither level stops phone-verified accounts.

In practice, members who hold a role are not held by the verification level. Do not hand out roles on join if you rely on it.

Step 2: Add captcha verification

A captcha raises the cost of each automated account: something has to solve it before the account can post. With Security Bot, new members get an Unverified role and see only the verify channel until they pass. They solve an image captcha in Discord, or pass the web check, which denies automated browsers. Follow the verification setup guide for the exact commands.

Step 3: Detect and stop join waves

Discord's own Raid Protection (Server Settings > Safety Setup > Raid Protection and CAPTCHA) detects join raids. It sends Activity Alerts to a channel you pick and as push notifications, and asks new joiners for a CAPTCHA for the next hour. It does not use your own scoring, punish the accounts that joined, or lock channels.

On Premium, Security Bot's anti-raid adds those parts. Each join scores points for a young account, a join seconds after the previous one, and a missing avatar. A join that matches none of them adds only the normal-join score.

When the total reaches your limit, anti-raid times out or bans the accounts that joined in that window and locks the channels you chose. A score limit of 20 to 30 with 3 points for a young account and 2 for a rapid join is a reasonable start. See how to protect your server from raids.

Step 4: Stop unauthorized bot applications

Anyone with Manage Server can add a bot application, and so can a phished moderator account that has it. Security Bot's Anti Bot Add trigger is free and on by default. It punishes whoever adds a bot and kicks that bot, unless they are the owner, ranked above Security Bot, or on the Bot-Add whitelist. Keep the trigger on and limit Manage Server to as few roles as possible.

Step 5: Filter what gets through

Some accounts will pass everything: aged, phone-verified, operated by a person for the first minute. Filters catch what they post. Discord's AutoMod has Mention Spam and Spam Content rules, and Community servers get Discord's suspicious-link filter.

On Premium, Security Bot's AutoMod page sets up those rules for you. It adds Threat Link Guard for scam and phishing domains, with a timeout, kick, or ban as the follow-up. See the anti-link guide for link filtering in detail.

Step 6: Pause invites when you are being hit

Server Settings > Invites > Pause Invites (needs Manage Server) disables every invite link, including the vanity URL, until you turn invites back on. For a timed pause, open the server name menu and use Security Actions > Pause Invites (up to 24 hours), which also offers Pause DMs.

Use it the moment a wave starts, set up the steps above, then resume. It is the only setting on this list that also blocks real members, so treat it as temporary.

What about anti-bot verification?

The phrase usually refers to captcha verification: proving a person is present. Security Bot's image captcha, web check, and Adaptive Verification are anti-bot verification. It does not refer to Discord's own verification level, which checks email, phone, and account age.

  1. Raise the verification level

    In Server Settings > Safety Setup (DM and Spam Protection on a Community server), set the Verification Level to Medium or High. Medium holds back accounts younger than five minutes or without a verified email. Accounts with a verified phone skip every level.

  2. Turn on captcha verification

    Invite Security Bot, run the setup commands, and choose Adaptive, Captcha code, or Web check. Accounts that do not pass stay in the verify channel.

    Command: /verification enable

  3. Detect join waves

    Keep Discord's Raid Protection on (Server Settings > Safety Setup > Raid Protection and CAPTCHA). On Premium, also turn on account age, join row, and no-profile-picture scoring with lockdown on Security Bot's Anti-Raid page.

  4. Block unauthorized bot additions

    Anti Bot Add is free and on by default in anti-nuke. Whoever adds a bot application is punished and the bot is kicked, unless they are the owner, ranked above Security Bot, or whitelisted. If it was turned off, turn it back on.

    Command: /enable action:Adding Bots

  5. Filter the messages that get through

    Set up Discord's AutoMod Mention Spam and Spam Content rules, or on Premium enable the matching presets plus Threat Link Guard on Security Bot's AutoMod page.

  6. Pause invites during an attack

    Server Settings > Invites > Pause Invites stops every invite link until you resume them. The server menu's Security Actions can also pause invites or DMs for up to 24 hours.

Which setting counters which bot

ThreatCountered by
Freshly created spam accountsVerification level Medium
Aged bot accounts joining one at a timeCaptcha verification
Hundreds of accounts joining at onceDiscord Raid Protection; anti-raid scoring and lockdown on Premium
A moderator adding a malicious bot applicationAnti Bot Add trigger
Human-operated accounts posting scam linksDiscord's suspicious-link filter and AutoMod; Threat Link Guard on Premium
An ongoing attack while you configure the restPause Invites

Keep Anti Bot Add on whatever your server's size. It is free, and there is no server size at which a moderator quietly adding an unknown bot application is a good idea. The setup guide covers the anti-nuke limits.

Common mistakes that let bots back in

  • Handing out a role on join. In practice any role skips the Discord verification level, and a role that is allowed into channels also overrides the Unverified deny. Let Security Bot give the Unverified role and nothing else until the member verifies.
  • Channels that are not synced with their category. A channel copies its category's permissions only while it is synced. A channel with its own allow for @everyone stays visible to unverified accounts even if the category denies it. Check unsynced channels, and run /verification update_role after adding channels.
  • Reusing a leaked invite. If a raid community has your link, every new wave uses it. Delete it under Server Settings > Invites, create a new one, and give links you post publicly an expiry or a maximum number of uses.
  • No security log channel. Anti-nuke warnings and punishments, the anti-raid lock, and web check denials are posted there. Verification passes are not logged, so test the gate with an alt rather than waiting for log entries. Assign the channel on the Logging page.
  • Whitelisting around verification. The whitelist covers anti-nuke triggers (including the invite-link filter) and, if you allow it, Beast Mode. It does not skip verification or join scoring. Use the account-age bypass in anti-raid, and /verification verify for one person. To admit only listed users, switch verification to Allowlist.
  • Lowering the score limit and forgetting it. A limit of 10 during an attack is right. A month later it flags a group of friends joining together. Go back to your usual limit once things are quiet.

If bot accounts are joining right now

  1. Pause invites in Server Settings > Invites, or with Security Actions in the server menu. The wave stops growing.
  2. Run `/lockall`. It locks every text channel the bot can manage, so nothing that already got in can post while you work. On Premium, anti-raid locks the channels on its list automatically once lockdown is on.
  3. Tighten verification. Set Adaptive to Strict, or switch to the web check with VPN blocking. The change takes effect without rebuilding the role or channel.
  4. Lower the anti-raid score limit (10 to 15) on Premium so the wave is timed out or banned sooner.
  5. Check the logs. Anti-raid timeouts and channel locks appear in the moderation log with Security Bot as the moderator and the reason "Anti Raid". Anti-raid bans are not logged by the bot. Discord's audit log shows them with the same reason. Accounts that got through can be selected on the Members page and banned together, or removed with /ban.
  6. Unlock and resume invites once joins have stopped. Run /unlockall, replace the leaked invite, and keep the tighter settings for a day.

Not set up yet?

If Security Bot is not in the server yet: invite it, move its role to the top, run /lockall, then enable verification. Anti-nuke starts acting as soon as the bot is in with Administrator and its role is above the accounts it watches.

Do this with Security Bot

Discord Verification Bot: Gate new members with Adaptive Verification, an image captcha, a no-click web check, one button, a passcode, or an allowlist.

Security Bot is a Discord verification bot that keeps new members behind an Unverified role until they pass. Adaptive Verification picks the check per member: low-risk accounts get straight in, others get an image captcha or a 5-second web check. When joins spike, it raises the bar on its own. Passing removes the role and can add your Verified role.

Frequently asked questions

Why do bots keep joining my Discord server?

Your invite link is findable and joining costs the attacker nothing. Make joining useless with verification and join protection instead of trying to hide the link.

Does the Discord verification level stop bots?

It holds back accounts without a verified email or younger than five minutes, and High adds a ten-minute wait in the server. Accounts with a verified phone number skip every level, which is why a captcha helps as well.

How do I stop someone from adding bots to my server?

Limit the Manage Server permission and keep Security Bot's Anti Bot Add trigger on. It punishes anyone who adds a bot application, except the owner, members ranked above Security Bot, and whitelisted users, and kicks the added bot.

Can I stop bots without a bot of my own?

Partly: the verification level, Rules Screening, Raid Protection (which adds a CAPTCHA for an hour after a detected raid), AutoMod, and Pause Invites are built in. A captcha on every join and join scoring with your own weights need a bot.

What is anti-bot verification?

A check that a new member has to pass before they can post. Security Bot provides it free as an image captcha, a web check that denies automated browsers, or Adaptive Verification, which picks one per member.

Do I need Premium to stop bots?

No. The verification level, Discord's Raid Protection and AutoMod, and Security Bot's verification (including Adaptive and the web check) and Anti Bot Add are free. Score-based anti-raid with lockdown, Security Bot's AutoMod presets, and Threat Link Guard are part of Premium, from $4.99 per month per server.

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.