Discord AutoMod Setup Guide 2026: Block Spam Before It Happens

Discord AutoMod is one of the most underutilized features in server management. It's a free, built-in system that runs 24/7, blocks harmful content before members see it, and generates logs you can review later.

In 2026, every server should have AutoMod enabled. Here's how to set it up correctly.

What Discord AutoMod Does#

AutoMod intercepts messages before they're posted. When a message triggers a rule, AutoMod can:

• Delete the message silently (user doesn't know why)
• Block it and send a warning (user sees a popup explaining why)
• Timeout the user (mute them temporarily)
• Send to mod review (flags it in a moderation channel for human review)

This happens in milliseconds. Most members won't even realize AutoMod exists - they just see fewer harmful messages in their server.

Accessing AutoMod Settings#

1. Open your server settings (click the server name at the top, then Server Settings)
2. Navigate to Safety Setup in the left sidebar
3. Click AutoMod to access the configuration panel

You need the Manage Server permission to configure AutoMod.

AutoMod Rule Types and Recommended Setup#

1. Spam Detection#

This rule automatically detects when users spam messages, mentions, or reactions.

Recommended settings:

• Trigger: Discord's default spam detection
• Response action: Timeout for 10 minutes (gentle warning)
• Exempt roles: Your Staff role (so mods can spam without getting timed out)
• When to increase: If you're still seeing spam, bump timeout to 1 hour and check exemptions

What it catches:

• Multiple messages posted in rapid succession
• Repeated identical messages
• Mention spam (@mention @mention @mention)
• Emoji spam
• Reaction spam

This rule alone stops most bot raids and spammers.

2. Custom Keywords#

This is the rule you customize for your community. Block words/phrases specific to your server's needs.

How to set it up:

1. Click Create Rule and select Block Custom Keywords
2. Add your keywords (one per line):
   • General category: slurs, harassment terms
   • Server-specific: spam phrases you've seen before, inside jokes you want to stop
   • Phishing domains: telegram.gg, discord-gift.xyz (fake Discord gift sites)
3. Choose response action: Delete + send warning (recommended)
4. Exempt roles: Staff (so mods can discuss moderation)
5. Exempt channels: Your staff channel or discussion channel where you might need to reference these terms

Building your keyword list:

• Start with 10-15 common terms
• After 2 weeks, review your mod logs and add terms you're manually deleting
• Don't go overboard - overly broad terms will create false positives
• Example: don't block "kill" if your gaming server uses it naturally ("kill that dragon")

Advanced technique: Test new keywords in a private channel before adding them to your live rule. This prevents accidental filtering of legitimate messages.

3. Link and Invite Blocking#

Phishing and malicious links are a constant threat. AutoMod can block them automatically.

Basic setup (safest for most servers):

1. Create a Block Links rule
2. Action: Block all links
3. Exempt roles: Staff only
4. Exempt channels: #media, #resources, #self-promotion (wherever links are expected)

This prevents random members from dropping suspicious links in #general but allows them in appropriate channels.

Intermediate setup (for larger/more active servers):

1. Block all links except approved domains
2. Whitelist trusted sites: your website, YouTube, github.com, etc.
3. Everything else gets deleted with a warning

Advanced setup (for servers with lots of legitimate links):

1. Allow links everywhere from Trusted Member role
2. Block all other links, but don't delete - send to mod review channel
3. Mods approve good links in the mod channel, and mods manually post them

4. Mention Spam Prevention#

Limit how many times a user can @mention other members in one message.

Recommended settings:

• Trigger: 5+ @mentions in one message
• Response action: Timeout for 1 hour
• Exempt roles: Staff (so mods can organize people)

This stops harassment and raids that target individuals with mass mentions.

5. New Account Filtering#

New, throwaway accounts are often used for raids and spam. Flag them.

Recommended settings:

• Trigger: Account younger than 24 hours
• Response action: Timeout for 10 minutes OR require verification (react to a message to proceed)
• Channels to apply: High-risk channels (#general, #announcements) or leave global
• Exemptions: Trusted roles (friends you know will join)

This rule doesn't block new members - it slows them down and flags them for human review.

Integration with Your Moderation Workflow#

AutoMod works best when paired with human moderation. Here's the workflow:

1. Set AutoMod to Log Flagged Content#

For soft-flag rules (instead of delete), set Send to mod review channel as the action. Create a dedicated channel like #mod-queue where AutoMod sends flagged messages for human review.

Advantages:

• Your mods see what AutoMod flagged
• You don't accidentally delete legitimate messages
• You can retrain AutoMod rules based on false positives

2. Review Your Rules Weekly#

1. Check your mod logs (Server Settings → Moderation)
2. See what AutoMod is catching
3. Identify gaps: "I see players getting reported for X that AutoMod misses"
4. Add new rules based on real patterns

3. Combine AutoMod with Bot Moderation#

AutoMod alone isn't enough. Pair it with a moderation bot (MEE6, Dyno, Probot) that provides:

• Persistent action history and appeal system
• Custom commands for manual moderation
• Logging to a dedicated channel
• Reputation systems (warn/mute/ban escalation)

Ideal setup:

• AutoMod: Instant spam blocking, obvious content filtering, new account flagging
• Moderation Bot: Warnings, mutes, bans, user history tracking
• Human Mods: Context-dependent decisions, appeals, culture shaping

Common AutoMod Mistakes and How to Fix Them#

Mistake 1: Overly Broad Keywords

• Example: Blocking "bad" when members say "that's bad ass"
• Fix: Use exact phrases or context-aware keywords. Test in a private channel first.

Mistake 2: Not Exempting Staff Channels

• AutoMod blocks staff from discussing moderation in private channels
• Fix: Exempt your #mod-only or #staff-discussion channel from most rules

Mistake 3: Blocking Legitimate Expression

• Example: Gaming communities need to say "kill," "attack," "spam" (as in spam abilities)
• Fix: Whitelist contexts. Make your keyword list server-specific, not generic

Mistake 4: Relying on AutoMod Alone

• AutoMod handles volume but can't judge intent
• Fix: Always have human mods on your team. AutoMod is a tool, not a replacement.

Mistake 5: Setting Actions Too Harsh

• Deleting messages silently without warning creates confusion
• Fix: Start with warnings (delete + send DM to user). Escalate to timeouts after warnings.

Testing Your Rules#

Before rolling out a new rule, test it:

1. Create a private channel (#automod-test)
2. Give yourself the "test user" role that's NOT exempt from the rule
3. Post messages that should trigger the rule
4. Verify the rule works as expected
5. Remove the test role and deploy the rule

AutoMod Best Practices#

1. Start conservative: Enable spam detection and link blocking. Add keywords gradually.
2. Monitor logs: Check what AutoMod is doing weekly. Refine rules based on real data.
3. Combine with humans: AutoMod handles 80% of spam. Your mods handle the other 20%.
4. Document your rules: In a staff channel, explain why each rule exists.
5. Communicate with members: In #rules, explain your AutoMod approach. Members will understand better if they know a rule exists.
6. Regular review: Every month, revisit your rules. Remove rules that aren't catching anything. Add rules for new patterns.

AutoMod Limitations to Know#

AutoMod can't:

• Detect context or sarcasm
• Handle rapid-fire coordinated attacks (too fast to react)
• Moderate voice channels
• Create custom action sequences (warn → mute → ban)
• Track repeat offenders across incidents

For these, you need human mods and a bot moderation system.

Your First AutoMod Setup (Quick Start)#

If you're starting from scratch:

1. Enable Spam Detection (1 minute)

   • Action: Timeout 10 minutes
   • Done.

2. Block Links (2 minutes)

   • Block all links
   • Exempt: Staff role, #media channel
   • Done.

3. Custom Keywords (5 minutes)

   • Add 5-10 baseline terms (common slurs, spam phrases)
   • Action: Delete + warn
   • Exempt: Staff
   • Done.

4. Mention Spam (1 minute)

   • Limit: 5 mentions
   • Action: Timeout 1 hour
   • Done.

That's a complete, functional AutoMod setup in 10 minutes. Expand from there based on what you observe.

AutoMod won't solve all your moderation problems, but it will handle the obvious, repetitive ones - freeing your mods to focus on the nuanced situations that actually need human judgment.