Auto-Moderation
Spam, caps, mention, link, zalgo, emoji, and duplicate-message filters, plus a banned word list and dry-run testing.
On this page
Overview
Auto-mod runs a set of independently toggleable rules against every message in real time: spam/mass-mention detection, excessive caps, link and invite filtering, zalgo (glitch text), emoji spam, duplicate/repeated messages, and a banned word list with severity tiers.
Turn on dry-run mode while you're tuning thresholds. Java Lava will log what it would have done to your mod log without warning, deleting, or timing anyone out, so you can catch false positives safely before going live.
Trusted roles and whitelisted channels are exempt from every rule globally, which is useful for staff channels or bot-command channels where normal message patterns don't apply.
Setup
- 1
Enable Auto-Moderation
Toggle the module on from Overview or the Moderation page's Auto-mod tab.
- 2
Turn on individual rules
Enable spam, caps, link, zalgo, emoji, and duplicate-message filters one at a time, each with its own thresholds and action (delete, delete+warn, delete+mute, delete+kick).
- 3
Build your banned word list
Add words or phrases to auto-delete, and set a default severity (auto-delete, auto-delete + warn, or auto-delete + mute) plus an optional regex mode for pattern matching.
- 4
Set exemptions
Add trusted roles and whitelisted channels that bypass every rule, and configure a minimum account age / server tenure for members to be treated as trusted.
- 5
Test with dry-run mode
Enable dry-run to log would-be actions to your mod log without taking them, then disable it once thresholds look right.
Examples
Illustrative mockups of what auto-moderation posts in Discord. They are not a live capture, but they match the real thing.
An auto-mod warning after the link filter catches an invite link
⚠️ Auto-Mod: Link Filter
Your message in #general was removed for containing a Discord invite link.
Member
@newbie42
Rule
Link / Invite Filter
Action Taken
Message deleted, warning issued
Dry-run mode is off, so this action was applied
Configuration options
| Option | Description |
|---|---|
Dry-Run Mode Default: Off | Logs what auto-mod would have done without deleting, warning, muting, or kicking. Safe for tuning thresholds. |
Spam / Mass-Mention Filter Default: On, 5 mentions, mute | Flags messages exceeding a configurable max-mentions-per-message threshold. |
Caps Filter Default: On, 70% caps, 10 char minimum, warn | Flags messages that are excessively uppercase, with a percentage threshold and minimum message length before it applies. |
Link / Invite Filter Default: On, blocks invites, delete | Blocks Discord invite links and/or general URLs, with a domain whitelist for exceptions (e.g. youtube.com). |
Zalgo Filter Default: On, delete | Blocks corrupted/glitch unicode text. |
Emoji Spam Filter Default: Off, 10 emoji max, warn | Flags messages with more than a set number of emoji. |
Duplicate Message Filter Default: On, 30s window, 3 duplicates, warn | Flags a member repeating the same message multiple times within a time window. |
Banned Words List | Custom list of words/phrases that are auto-deleted on match, with per-severity actions (delete / delete+warn / delete+mute) and an optional regex mode. |
Trusted Roles Default: Moderator, Admin, Trusted | Roles exempt from every auto-mod rule. |
Whitelisted Channels Default: #staff-chat, #bot-commands | Channels exempt from every auto-mod rule. |
Trusted Account Age / Server Tenure Default: 7 days account age, 3 days tenure | Minimum account age and server membership duration (in days) before a member is treated as trusted for exemption purposes. |
Permissions
Bot permissions required
Who can configure this
Anyone with the dashboard's Owner or Admin access tier (Staff Access page), or Editor tier for day-to-day config changes