Message Filtering Commands

Advanced content filtering to keep your server safe and clean.

Filter Severity Levels

The filter has three levels of severity that determine how violations are handled:

Minor - Logs the message in mod logs only
Moderate - Deletes the message and flags an automod warning
Severe - Deletes the message, flags a warning, and mutes the user for 5 hours

Automatic Actions

Automatic Actions allow you to take consistent, escalating action against users who repeatedly trigger filters. These are managed via the Automod & Filters dashboard.

Key Concepts

Trigger Level: The number of times a user must trigger a filter before the action is taken.
Scaling Multiplier: Multiplies the punishment duration on repeat offenses. For example, a 10m mute with a 2.0x multiplier will become 20m, then 40m, etc.
Re-offense Window: The period of time the bot "remembers" the previous violation for scaling purposes.
Execution Lock: To prevent "scaling storms," the bot has a 10-second lock per user. This ensures that a single burst of 10 messages only triggers one auto-action.

Managing Rule

Moderators can view a user's current trigger history and reset their scaling level using the Member Automod Status lookup on the dashboard or via the /automod-member command.

Word Filter

filter word add

Adds a word to the message filter.

Usage:

/filter word add [word] [severity: Minor/Moderate/Severe]

Example:

/filter word add dynobot Minor

Severity Options:

Minor - Logs only, message stays
Moderate - Deletes message, warns user
Severe - Deletes, warns, mutes for 5 hours

Required Role: Moderator

Multiple Words

You can add multiple variations of words to catch similar phrases.

filter word list

Generates a complete list of all currently filtered words.

Usage:

/filter word list

Output:

Shows all filtered words
Displays severity level for each word
Sent to the moderator log channel

Required Role: Moderator

Advanced Word Filter Protections

Configure how aggressive the word filter should be. These settings are managed via the web dashboard.

Feature	Description
Leetspeak	Detects character substitutions (e.g., `f*ck`, `@ss`)
Homoglyphs	Normalizes look-alike characters from other alphabets
Separators	Detects words with symbols between letters (e.g., `s.h.i.t`)
Duplicate Letters	Catches stretched words (e.g., `caaaat` -> `cat`)
Headers	Blocks messages using markdown headers (e.g., `# Header`)
Subtext	Blocks messages using markdown subtext (e.g., `-# Small text`)
Masked Links	Blocks masked markdown links (e.g., `[text](url)`)
Workarounds	Extreme normalization to catch `w o r d` style bypasses

Name Moderation

Moderates usernames, display names, and nicknames against the server's word filter. These settings are managed via the Automod & Filters dashboard.

Toggles

Username Moderation: Checks global Discord usernames.
Display Name Moderation: Checks global Discord display names.
Nickname Moderation: Checks server-specific nicknames.

Configuration

Actions can be configured for when a bad name is detected:

Log only: Violations are only logged to the moderator log channel.
Warn: Users are issued a formal warning and logged.
Auto-change nickname: The bot attempts to set a compliant server nickname, DMs the user (if it was a nickname violation), and logs the change.
Mute: Users are muted. Once they change their name to a compliant value, they are automatically unmuted.

Auto-Unmute

The auto-unmute feature only triggers if the user was muted specifically for a name violation. If the user was muted manually by a moderator or for another reason, they will remain muted.

Invite Filter

Advanced configurations can be enabled on the dashboard:

Separators Protection: Catch invite links hidden with dots or spaces (e.g. d.i.s.c.o.r.d.gg/abc).
Workaround Detection: Strips all non-alphanumeric characters to catch heavily obfuscated invite links.
URL Decoding Protection: Decodes URL encoding before checking for invites.

filter invite toggle

Turns the invite link filter on or off.

Usage:

/filter invite toggle

Effect:

Prevents users from posting Discord invite links (discord.gg, discord.com/invite)
Prevents posting links to major server listing sites (such as Disboard)
Helps prevent server poaching and unauthorized promotions

Required Role: Moderator

filter invite whitelist-add

Adds an invite link or server to the whitelist (allow list).

Usage:

/filter invite whitelist-add [invite]

Example:

/filter invite whitelist-add disboard.org/server/example

Notes:

Whitelisted invites will NOT be filtered
Useful for official community servers or partner servers

Required Role: Moderator

filter invite whitelist-remove

Removes an invite from the whitelist.

Usage:

/filter invite whitelist-remove [invite]

Example:

/filter invite whitelist-remove discord.gg/example

Required Role: Moderator

filter invite whitelist-list

Lists all whitelisted invites.

Usage:

/filter invite whitelist-list

Output:

Shows all whitelisted invite codes/links
Sent to the moderator log channel

Required Role: Moderator

filter invite server-whitelist-add

Adds a server ID to the whitelist. Any invite link that resolves to this server (no matter the specific code) will be allowed.

Usage:

/filter invite server-whitelist-add [server_id] [nickname: optional]

Example:

/filter invite server-whitelist-add 1178595123408424990 Partner Server

Notes:

server_id must be a numeric Discord snowflake
nickname is a friendly name for your own organization on the dashboard/list
The bot resolves invite links to check their target server ID against this list

Required Role: Moderator

filter invite server-whitelist-remove

Removes a server ID from the whitelist.

Usage:

/filter invite server-whitelist-remove [server_id]

Example:

/filter invite server-whitelist-remove 1178595123408424990

Required Role: Moderator

filter invite server-whitelist-list

Lists all whitelisted server IDs and their friendly nicknames.

Usage:

/filter invite server-whitelist-list

Output:

Shows all whitelisted server IDs and nicknames
Sent to the moderator log channel

Required Role: Moderator

Smart Filtering

Intelligent content filtering using AI/ML models.

filter smart configure

Enables or disables smart content filters.

Usage:

/filter smart configure [type]

Available Types:

Adult Content - Filters NSFW/adult content
Malware - Filters potential malware links
Phishing - Filters phishing/scam links
Gambling - Filters online gambling links
Piracy Content - Filters links to piracy and illegal copyright-sharing domains
Third-Party Clients - Blocks links to third-party Discord clients/reimplementations
Third-Party Mods - Blocks links to ToS-breaking client modifications and plugins
Video Exploits - Detects and blocks spoofed duration video uploads (infinite loop glitches)

Example:

/filter smart configure Adult Content

Notes:

Uses advanced detection algorithms
Can be toggled on/off as needed

Required Role: Moderator

filter smart nsfw-exception

Allows adult content in NSFW channels even when the adult filter is enabled.

Usage:

/filter smart nsfw-exception

Effect:

Toggles NSFW channel exception
When enabled, NSFW channels bypass adult content filter
Useful if you have designated adult content areas

Required Role: Moderator

Organization

Use NSFW channels for adult content rather than filtering it across the entire server.

GIF Filter

When enabled, all GIFs (Tenor, Giphy, direct links, and file uploads) will be automatically deleted. Staff members are exempt. This is managed via the Automod & Filters dashboard.

Link Filtering

Advanced control over URLs in your server. These settings are managed via the Automod & Filters dashboard.

Prevent All Links

When enabled, every URL posted by a non-staff user will be deleted, unless it contains a string found in the Link Whitelist.

Block Specific Links

When "Prevent All Links" is disabled, you can still block specific domains or URLs by adding them to the Link Blocklist.

Management

Link Whitelist: Add strings (domains or URL segments) that should be allowed when "Prevent All Links" is ON.
Link Blocklist: Add strings (domains or URL segments) that should be blocked when "Block Specific Links" is ON.

Bad Image Detection

Detects known scam or malicious images using perceptual hashing. This is managed via the Automod & Filters dashboard.

Configuration

Moderation Action: Choose between Log Only, Warn, Mute, Kick, or Ban when a blocked image is detected.
Blocked Image Hashes: Manage the list of perceptual hashes to block.
Labels: Categorize hashes (e.g., "Crypto Scam") for better organization.

Image Hash Creator

Upload an image to the dashboard to calculate its perceptual hash and add it to the blocklist in one click.

Image Spam Protection

Take automated action if a user spams blocked images across multiple channels in a short period.

Threshold: The number of bad images allowed before action is taken.
Window: The time window (in seconds) for the threshold check.
Action: The automated punishment to execute.

Emoji & Sticker Filters

filter emoji toggle

Toggles filtering of external (animated/custom) emojis.

Usage:

/filter emoji toggle

Effect:

When enabled, prevents posting of emojis not from your server
Reduces clutter from external emoji spam
Users can still use server emojis

Required Role: Moderator

filter sticker toggle

Toggles filtering of external stickers.

Usage:

/filter sticker toggle

Effect:

When enabled, prevents posting of stickers not from your server
Helps maintain a clean sticker experience
Users can still use server stickers

Required Role: Moderator

This will create a custom AutoMod rule in your Discord server to prevent it before the message goes through. The benefit of using Raptor to manage this over creating it yorself, is it will still use Raptor's Auto Punishment system and count as a Raptor Infration and still will work with Raptor Automod Actions for spammers.

Required Role: Moderator

filter mention add-user

Blocks mentions of a specific user.

Usage:

/filter mention add-user [user]

Example:

/filter mention add-user @Wumpus

Required Role: Moderator

filter mention remove-user

Removes a user from the mention blocklist.

Usage:

/filter mention remove-user [user]

Required Role: Moderator

filter mention add-role

Blocks mentions of a specific role (e.g., prevents pinging @Developer).

Usage:

/filter mention add-role [role]

Required Role: Moderator

filter mention remove-role

Removes a role from the mention blocklist.

Usage:

/filter mention remove-role [role]

Required Role: Moderator

filter mention add-member-role

Blocks mentions of any user who holds a specific role. This is useful for protecting all staff members or specific groups from being pinged.

Usage:

/filter mention add-member-role [role]

Example:

/filter mention add-member-role @Staff

Required Role: Moderator

filter mention remove-member-role

Removes a role from the protected member-role list.

Usage:

/filter mention remove-member-role [role]

Required Role: Moderator

filter mention list

Displays a list of all currently blocked users, roles, and protected roles.

Usage:

/filter mention list

Required Role: Moderator

Voice Moderation

Configures automatic disconnection for users engaging in specific voice activities.

Technical Note

These features provide granular control because Discord's native permission system does not allow separating "Go Live" (streaming) and "Camera" (video) privileges, as both are bundled under the single "Video" permission.

filter voice camera

Toggles the automatic disconnection of users who turn on their camera in a voice channel.

Usage:

/filter voice camera

Required Role: Administrator

filter voice golive

Toggles the automatic disconnection of users who start a "Go Live" stream in a voice channel.

Usage:

/filter voice golive

Required Role: Administrator

Note: Toggles are also available on the Web Dashboard under Automod & Filters.

Smart Filters

Configure domain blocklists and specialized message filtering dynamically. Managed via the Web Dashboard under Automod & Filters.

filter smart configure

Description: Toggles a smart filter category.

Usage:

/filter smart configure [type]

Choices:

Adult Content: Blocks URLs from known adult websites.
Malware: Blocks domains known for malware, trackers, and malicious scripts.
Phishing: Blocks known phishing, scamming, and credentials-harvesting websites.
Gambling: Blocks domains known for online gambling and betting.
Piracy Content: Blocks domains known for piracy and illegal sharing of copyrighted content.
Suicide Promotion: Blocks links to websites and communities that promote self-harm or suicide methods. Crisis lines, suicide hotlines, and prevention resources are still allowed.
Drug Forums: Blocks links to online drug forums and narcotics-related domains.
Client Modifications: Blocks domains promoting client modifications violating Discord ToS (BetterDiscord, Vencord, etc.).
Third-Party Clients: Blocks stand-alone alternative Discord client domains.
Rich Presence Invites: Blocks interactive rich invites like Spotify "Listen Along" presence objects or game lobby joins.
Video Exploits: Detects and blocks spoofed duration video uploads (infinite loop glitches).

Read-Only Mode

filter readonly toggle

Toggles server-wide read-only mode for non-staff users.

Usage:

/filter readonly toggle

Effect:

Deletes non-staff messages while enabled
Deletes newly created non-staff threads
Removes non-staff reactions as they are added
Disconnects non-staff users who join voice channels

Required Role: Moderator

Automod Debug Mode

filter debug-mode

Toggles automod debug mode for staff testing.

Usage:

/filter debug-mode

Effect:

Allows moderators/admins/owners to intentionally trigger filter enforcement for testing
Temporarily disables normal staff exemption for filter actions
Keeps read-only mode staff exemption unchanged

Required Role: Moderator

Level-Based Whitelist

Allow loyal members to bypass specific filter and anti-spam checks based on their leveling rank. This allows you to reward trusted users with fewer restrictions while maintaining tight security for new members.

whitelist set

Sets a minimum level requirement for a feature bypass.

Usage:

/whitelist set [feature] [level]

Example:

/whitelist set Word Filter 5

Configurable Features:

Filters: Word, Invite, Emoji, Sticker, Mentions
Smart Filters: Adult Content, Malware, Phishing, Gambling, Piracy, Suicide Promotion, Drug Forums
Anti-Spam: Message Rate, Duplicates, Mentions, Links, Attachments, Emojis, Newlines, Characters, Caps, Unicode

Note: Set level to 0 to disable the bypass for that feature.

Required Role: Admin

whitelist view

Displays all currently configured level bypass thresholds.

Usage:

/whitelist view

Required Role: Moderator

Filter Management Tips

Start Broad - Begin with severe violations
Monitor Logs - Check what gets filtered
Adjust Severity - Modify severity levels based on false positives
Whitelist Carefully - Only whitelist trusted invites
Communicate - Let your community know your filter rules

False Positives

Monitor your filtered words list for innocent words that might get caught. For example, common words in names or slang.