Paladin — Documentation

Paladin is a 24/7 Discord server-protection and moderation bot: anti-raid, anti-spam, scam-link blocking, content filtering, and more — set up in minutes.

Last updated: 2026-06-28

Installation

Step 1 — Add the bot. Open this link in any browser, or click it:

Add Paladin to your server

Step 2 — Pick your server. Choose a server you have permission to manage, and authorize Paladin. The link grants Administrator, so Paladin works fully the moment it's added — nothing else to configure.

Step 3 — Set Paladin's role. In Server Settings → Roles, drag Paladin's role up so it sits above the members it needs to act on (mod/manager level). Even with Administrator, Paladin can only mute, quarantine, or ban members whose highest role is below its own.

Step 4 — You're protected. Run /paladin to see every command. Paladin automatically starts protecting all channels (or only the ones you choose in settings).

Plans & pricing

Paladin is free for any server under 250 members — full access to every feature, no subscription.

Once a server passes 250 members, it keeps the core protection free and unlocks the full feature set with a per-server subscription:

The subscription is a server (guild) subscription — bought for one specific server by someone who manages it, and billed entirely through Discord. It applies to that one server only. Check status or subscribe with /paladin premium in your server.

Commands

Every command lives under one command: /paladin. Just type /paladin in any channel to see them all. (Examples: /paladin mute @user, /paladin protect on.)

Moderation

Raid protection

Raid protection is on by default. Beyond the burst trigger above, it also watches for slow-burn raids — a sustained trickle of new-account joins over a longer window that wouldn't trip the per-window join count on its own. When a raid is detected, Paladin can put the server into lockdown, and that lockdown auto-lifts once the wave subsides (a manual /paladin protect on lockdown never auto-lifts). Floods at 3× the join threshold trip regardless of account age.

ChatSpeed

Anti-spam

Spam protection is on by default — it runs automatically (no per-rule setup) and catches:

• Cross-channel spam — the same content fired across multiple channels.
• Rapid duplicates — the same message repeated in quick succession.
• Floods — high-rate message bursts from one member.
• Coordinated multi-account spam — the same payload posted by several accounts together.
• Rate abuse — sending faster than a human plausibly types.
• Image copycats — the identical image posted 2+ times triggers a mute or ban.

Content filter

There is no separate /paladin rating on or /paladin rating off command — choosing a tier with /paladin rating set enables the filter, and /paladin rating set off disables it.

What each rating means

• G — strictest. Censors all ranks except the server owner.
• PG / PG13 / R — progressively more permissive profanity tiers.
• Custom — enforces only the words you add with /paladin badword (your custom blocklist).
• off — disables the profanity filter (the always-blocked categories below still apply).

Always blocked (cannot be disabled)

Regardless of rating tier — including off, Custom, and even for the server owner — Paladin always blocks:

• Racial, religious, and LGBTQ slurs.
• Hate speech and extremism (including violent-verb × protected-group incitement).
• Child-safety / CSAM content — blocked at every tier, for every user including the owner, and never allowlistable.

Evasion resistance

The matcher is hardened against the usual dodges — leetspeak (b4dw0rd), inserted spacing, accents/diacritics, and zero-width characters are all normalized before matching. It is false-positive-safe on innocent words that merely contain a blocked substring (e.g. Scunthorpe, analysis, Fukushima are not flagged). Multilingual profanity is covered for es, fr, de, pt, it, and sv. Sexual-solicitation detection is contextual — it looks at a vulgar verb next to a body-part term, so "suck a dik" is blocked while "dik" on its own is not.

Scam / phishing links

/paladin links action sets what happens on a confirmed scam link. The choices are quarantine, timeout, delete only, and ban.

Detection draws on a ~430,000-domain blocklist assembled from maintained public feeds (Discord's scam/phishing list plus the blocklistproject scam/phishing/fraud/malware host lists). Matching is done locally — Paladin only downloads the domain-name text and never visits or fetches the flagged sites. On top of the blocklist it also catches IP-grabbers, masked-link mismatches, homoglyph/IDN look-alikes, and URL shorteners.

A scam/explicit keyword heuristic runs on both the domain and the path, and it fires even on allowlisted hosts. Allowlisting a host with /paladin links allow-add does not exempt scam or explicit URL paths on it — e.g. trusting discord.gg will not let discord.gg/freenudes through.

Paladin also suppresses non-staff link-preview images (on message create and update) when the image filter is on; mods and above are exempt.

AutoMod (native Discord rules)

Paladin maintains two Discord AutoMod rules for you under Server Settings → AutoMod:

• "Paladin . Word Filter" — profanity/sexual/slur presets mapped to your rating (G/PG block all; PG13 blocks sexual + slurs; R blocks slurs).
• "Paladin . Scam Links" — a domain blocklist sample drawn from the scam-link feeds.

These are Discord's own rules, so they block content before it is ever posted. Paladin re-syncs them automatically — on /paladin rating set, when scam-link protection is toggled, and once on startup — so you should not hand-edit or delete them; Paladin will overwrite manual changes on the next sync.

A few platform limits to be aware of (these are Discord's, not Paladin's):

• Discord allows only one preset-keyword AutoMod rule per server.
• AutoMod cannot exempt a single user, so even the server owner is not exempt from these native rules — though Paladin's own in-app filter still exempts the owner.
• These rules intentionally skip bots and webhooks (so sparring/training bots aren't caught).

Logs

Info

Who can use Paladin's commands

Paladin's commands are mod-only. The /paladin command is hidden from regular members and shown automatically to server administrators and the server owner. To let a specific moderator role use it, you have two easy options:

• Grant it in Integrations (one toggle): Server Settings → Integrations → Paladin → add your mod role under Roles & Members. Because everything is one command now, this is a single switch — no per-command setup.
• Or give the role the Manage Server permission: Server Settings → Roles → [your role] → enable Manage Server. Anyone with it will see /paladin.

Tip: after changing who can use it, have that member fully reload Discord (Ctrl+R, or restart the app). Discord caches the slash-command list and won't refresh it live.

Note: the command tables above cover the server-admin commands. /paladin comp (whitelisting a server to bypass billing) is bot-owner-only — it is enforced by a hard owner check, so a server admin can never run it even with Administrator.

The bot's own permissions

Separate from who can use the commands, Paladin itself needs permissions to take action (mute, lock down, set filters). The install link grants Administrator, so a fresh install works immediately.

If Paladin ever reports a missing permission, fix it in Server Settings → Roles → Paladin → Permissions — enable Administrator (or at minimum Manage Server, Manage Channels, Manage Roles, Timeout Members, Ban Members), then Save Changes.

Re-inviting a bot that's already in your server does not update its existing permissions — Discord only applies them on the first add. So either edit the role as above, or kick and re-add Paladin with the install link.

Troubleshooting

• A mod can't see /paladin. Their role hasn't been granted yet — grant it in Integrations or give it Manage Server (see above) — and they should reload Discord (Ctrl+R). Admins always see it.
• "Couldn't change the image filter" on /paladin rating. If your server is a Community server, Discord keeps its explicit-image filter on and won't let it be turned off — that's expected, and your content rating still applies regardless.
• The new /paladin commands aren't showing yet. Global command updates can take up to ~1 hour to propagate across Discord. Reload Discord to refresh; they'll appear shortly.
• AutoMod rules named "Paladin . Word Filter" / "Paladin . Scam Links" keep reappearing or reverting my edits. That's expected — Paladin creates and maintains those two rules under Server Settings → AutoMod and re-syncs them automatically (on rating change, on scam-link toggle, and on startup). Don't hand-edit or delete them; any manual change is overwritten on the next sync. They block content before it's posted and intentionally skip bots/webhooks. Note Discord's own limits: only one preset-keyword rule per server, and AutoMod can't exempt a single user (so even the owner isn't exempt from these native rules).