# Toxicity Prevention

Our <mark style="color:blue;">**Toxicity Filter's**</mark> is designed to detect and filter out harmful content, including insults, identity-based attacks, and threats. It utilizes a customizable scoring system, giving yourself the flexibility to set thresholds for different severity levels. This ensures that only messages surpassing your chosen extreme rating trigger the filter, allowing for tailored moderation to suit your specific needs.

## Settings

Here you can choose all available settings for the <mark style="color:blue;">**TOXICITY FILTER**</mark>

Navigate to the Automod tab following these directions Dashboard > AUTOMOD > <mark style="color:blue;">**TOXICITY FILTER**</mark>. This will allow you to adjust the settings of the filter according to your specific requirements, including setting threshold levels and configuring the scoring system to better suit your community's tone and guidelines.

<figure><img src="/files/Xk06yoo2KwSjcHb4enj2" alt="" width="111"><figcaption></figcaption></figure>

<figure><img src="/files/3ETRMslLfjvXxLgpcaBK" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/TPPvahoS3wvAI2ynQNuP" alt=""><figcaption><p>If your toxicity filter looks like this, please refer to <a data-mention href="/pages/sG1rtL4CICVZtjklalMt#setting-my-quarantine-system">/pages/sG1rtL4CICVZtjklalMt#setting-my-quarantine-system</a></p></figcaption></figure>

<figure><img src="/files/vn2n33AhN6tyeq2QxOUd" alt=""><figcaption><p><strong>EXAMPLE PICTURE</strong></p></figcaption></figure>

### Scoring

Toxicity scoring functions by evaluating messages based on specific categories like insults, threats, or identity attacks. Each category is assigned a percentage score representing the level of toxicity. If a message scores above a defined threshold, such as 75% for insults, it will be flagged for moderation. This ensures that even if a message scores below a critical level for threats or identity attacks, it will still be moderated due to the high toxicity level in insults, enhancing the filter's effectiveness in maintaining a safe environment.

### Analysing Command

The `analyse` command allows you to evaluate text or images to determine their content score. If the analysed text receives a tick mark, it indicates that the automod will be triggered if enabled. Additionally, you receive a toxicity percentage score, which shows how toxic the content is on a scale of 0 to 100.

<figure><img src="/files/8GJEyiptRJM70t267r5T" alt="" width="485"><figcaption><p><strong>EXAMPLE IMAGE</strong></p></figcaption></figure>

## Automod in Action

Here is some brief examples of the auto-mod working below in a real server...

<figure><img src="/files/M7RnEUxhCdlzr4etaZ2h" alt="" width="490"><figcaption></figcaption></figure>

<figure><img src="/files/UXxcNRc1UtyqxxQl14iU" alt="" width="431"><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://guide.curvebot.xyz/auto-moderation/toxicity-prevention.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.