Sentinel

The Sentinel module (sentinel) is a Premium feature that uses Artificial Intelligence to detect and filter Hate Speech on your server, helping to maintain a safe and respectful community.

Configuration options

Option	Description
enabled	true if you want the bot to activate Sentinel.
maxLvl	The maximum authorization level affected by the filter.
threshold	The detection threshold (0.0 to 1.0). The higher the value, the harder it will be to trigger.
deleteMessages	true to delete offensive messages automatically.
validationChannel	ID of the channel for manual validation and system training.
validationLvl	Minimum level required to use manual validation.
ignoredRoles	IDs of roles that the bot will not analyze.
ignoredChannels	IDs of channels that the bot will not analyze.
language	Primary detection language (e.g. "es").

Validation and Thresholds

It is recommended to start with a high threshold (e.g. 0.85) and gradually lower it if you see too many offensive messages passing through, while using the validationChannel to adjust the bot’s accuracy manually.

Configuration example

sentinel:
  enabled: true
  maxLvl: 20
  threshold: 0.85
  deleteMessages: true
  validationChannel: '333333333333333333'
  validationLvl: 50
  ignoredRoles: []
  ignoredChannels: ['444444444444444444']
  language: "es"