> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mnrl.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Content Classifier: Detect harmful content with AI

> Monorail AI Classifier analyzes text and categorizes it for potentially harmful content across multiple categories, helping you moderate content at scale.

The Classifier tool analyzes content and flags it as potentially harmful across several predefined categories. Use it to review user-submitted text, moderate content at scale, or screen inputs before publishing.

## Using the Classifier

<Steps>
  <Step title="Open Classifier in your dashboard">
    Click **Classifier** in the sidebar of your Monorail AI dashboard.
  </Step>

  <Step title="Enter your text">
    Paste or type the text you want to analyze into the input field.
  </Step>

  <Step title="Click Classify">
    Submit the text and let the AI evaluate it against the harm categories.
  </Step>

  <Step title="Review the results">
    Examine the category breakdown returned by the AI, which indicates potential harm levels across each dimension.
  </Step>
</Steps>

## Classification categories

The Classifier evaluates content across the following categories:

* Hate speech and discrimination
* Violence and graphic content
* Self-harm or dangerous activities
* Sexual content
* Spam or deceptive content

<Info>
  The Classifier provides probabilistic scores — it is an AI tool and may not catch all harmful content or may produce false positives. Always combine automated classification with human review for critical use cases.
</Info>

## Common use cases

* Screening user-submitted content on platforms
* Pre-publishing review for brand-safety compliance
* Research and content analysis

## Credit usage

Classifier uses text models and charges per token processed. The cost scales with the length of the content you submit. See the [Credits](/credits) page for current rates.