---
title: /imagefilter settings
description: Automatically detect and remove NSFW, nudity, or gore images.
---
## Overview

`/imagefilter settings` opens an interactive dashboard for scanning images posted in your server and removing ones that contain NSFW content, nudity, or gore. Image scanning is powered by a third-party API (JigsawStack) and requires your own API key.

## Required Permissions

- **Administrator**

## Subcommands

### settings

No options. Running the command opens a dashboard that walks you through setup, then exposes an Overview, Punishment, and Whitelist page once configured.

## Setup

The first time you run the command, the dashboard guides you through creating a JigsawStack account, generating an API key, and pasting it into BetterAntiSpam. The key is tested before it's saved, and rejected keys show the reason.

## Dashboard settings

Once a working API key is saved, the dashboard is split into three pages: Overview, Punishment, and Whitelist.

### Overview page

A master toggle turns image filtering on or off. A detection level (low, medium, or high) controls how strict the filter is, default medium.

### Punishment page

Choose the response to a flagged image: delete the message (default), warn, mute, kick, or ban.

- If mute is selected, enter a duration such as `10m` or `1h30m`, default 10 minutes.

### Whitelist page

Add channels, roles, or users exempt from image filtering.

## Behavior

- BetterAntiSpam tracks your API usage against your plan's quota and shows a usage bar in the dashboard.
- If your key becomes invalid or your quota runs out, the dashboard locks into a "paused" state until you provide a working key. BetterAntiSpam automatically rechecks your key's status periodically.
- A **Reset all** button restores detection and punishment settings to their defaults and clears all whitelists, after a confirmation step.
- Only the admin who opened the dashboard can interact with it.

<Banner variant="info">
This feature requires your own JigsawStack API key and is subject to that service's usage limits.
</Banner>