AutoMod: Block Commonly Flagged Words

What it does

This is disabled by default. When enabled, Discord can perform actions whenever someone sends a word or phrase that is commonly flagged as bad.

You can choose a mix of up to three different word filters:

Severe Profanity — extreme forms of swearing or cursing.
Insults & Slurs — personally insulting terms, and terms that may be considered hate speech.
Sexual Content — terms that refer to sexually explicit behavior or activity.

You can also add exceptions for words or phrases using Discord’s simple text-pattern syntax. You cannot add regex exceptions.

If a message is flagged, you can choose a combination of actions to perform:

Block the message from being sent.
Optionally show a small notice message to the user when blocked.
Send an alert to a pre-defined channel — for example, to notify staff.

You can also choose channels, categories, and roles where the rule is ignored.

Anyone with the Administrator or Manage Server permission is always ignored. If some of your staff don’t have these permissions, you may need to add their role(s) to the ignore list.

False positives

Depending on your community’s target audience, some of these filters may produce false positives if you have a lot of adults. There can be perfectly legitimate words for adults to use that may not be suitable around children.

Some words are more “innocent” than others, and users may come to see them as trigger words over time.

If you’re unsure what to do, you can enable everything to begin with and then gradually add exceptions you find acceptable, or even disable specific categories outright.

Another approach is to not block messages at the beginning, but enable alerting only — so you can tune the rules before they start blocking messages.