Agentic moderator

An automatic, AI-powered moderator that handles text and image content

Supported languages

Afrikaans, Albanian, Amharic, Arabic, Armenian, Assamese, Azerbaijani, Basque, Belarusian, Bengali, Bosnian, Bulgarian, Burmese, Catalan, Cebuano, Chichewa, Chinese, Corsican, Croatian, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Frisian, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Kinyarwanda, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Nyanja, Odia, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scots Gaelic, Serbian, Sesotho, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Tatar, Telugu, Thai, Turkish, Turkmen, Ukrainian, Urdu, Uyghur, Uzbek, Vietnamese, Welsh, Xhosa, Yiddish, Yoruba, Zulu

Performance may vary by language, especially for low-resource or morphologically complex languages.

The moderation model is designed to detect harmful content categories (like hate speech, violence, harassment, etc.) and may work even if the language is low-resource, due to its multilingual training.

automod

• Usage: !automod

• Checks: server_only

Automated AI moderation for chats, images, and emotes powered by the latest frontier moderation models.

automod threshold

• Usage: !automod threshold <threshold>

• Restricted to: ADMIN

Set the moderation threshold for message sensitivity. The threshold value should be between 0 and 1, where: 0.00 represents a very sensitive setting, capturing more messages for moderation. 1.00 represents a barely sensitive setting, allowing most messages (even ones containing explicit content) to pass through without moderation. Adjust this setting based on your community's needs for moderation sensitivity.

Recommendations

For general communities, a threshold of 0.50 is often effective. For professional communities (or if stricter moderation is preferred), consider a threshold below 0.40. For more lenient settings, a threshold above 0.70 might be suitable.

automod settings

• Usage: !automod settings

• Restricted to: ADMIN

Show the current settings of the cog.

automod reasons

• Usage: !automod reasons

Explains how the AI moderator labels and categorizes content

automod logs

• Usage: !automod logs <channel>

• Restricted to: ADMIN

Set the channel to log moderated messages.

automod disclaimer

• Usage: !automod disclaimer

• Restricted to: ADMIN

Toggle the monitoring warning message that is periodically sent to channels.

Disabling this warning is not recommended and may have legal or compliance implications. If you're using Sentri in a company-facing server, don't toggle this before talking with your legal team.

automod delete

• Usage: !automod delete

• Restricted to: ADMIN

Toggle whether violatory messages are deleted or not.

automod toggle

• Usage: !automod toggle

• Restricted to: ADMIN

Toggle automatic moderation on or off.

automod whitelist

• Usage: !automod whitelist

• Restricted to: ADMIN

Manage whitelisting/"bypassing" parts of the server from moderation.

automod whitelist channel

• Usage: !automod whitelist channel <channel>

Bypass/unbypass channels

automod whitelist category

• Usage: !automod whitelist category <category>

Bypass/unbypass a channel category

automod whitelist user

• Usage: !automod whitelist user <user>

Bypass/unbypass a user

automod whitelist nsfw

• Usage: !automod whitelist nsfw

Toggle NSFW bypass status

automod whitelist role

• Usage: !automod whitelist role <role>

Bypass/unbypass users in a role

automod history

• Usage: !automod history [user=None]

• Restricted to: ADMIN

Show the violation history for a user in this server. If no user is provided, shows your own history (if you are not a bot). Shows 5 violations per page, with emoji reactions to scroll if there are more. Also includes a graph of abuse trends over time.

automod stats

• Usage: !automod stats

Show statistics of the moderation activity.

automod vote

• Usage: !automod vote

• Cooldown: 1 per 86400.0 seconds

Give feedback on the server's agentic moderation

automod timeout

• Usage: !automod timeout <duration>

• Restricted to: ADMIN

Set the timeout duration in minutes (0 for no timeout).