---

name: humanize description: Transform AI-generated text into natural, human-like content that bypasses AI detectors like GPTZero, Turnitin, and Originality.ai. Uses credits based on word count. user-invocable: true argument-hint: "[text to humanize] [--intensity light|medium|aggressive]" allowed-tools: WebFetch

Humanize AI Text

Transform AI-generated content into natural, human-like writing using the HumanizerAI API.

How It Works

When the user invokes /humanize, you should:

1. Parse $ARGUMENTS for text and optional --intensity flag
2. Call the HumanizerAI API to humanize the text
3. Present the humanized text with before/after scores
4. Show remaining credits

Parsing Arguments

The user may provide:

• Just text: /humanize [their text]
• With intensity: /humanize --intensity aggressive [their text]

Default intensity is medium.

Intensity Levels

Value	Name	Description	Best For
light	Light	Subtle changes, preserves style	Already-edited text, low AI scores
medium	Medium	Balanced rewrites (default)	Most use cases
aggressive	Bypass	Maximum bypass mode	High AI scores, strict detectors

API Call

Make a POST request to https://humanizerai.com/api/v1/humanize:

Authorization: Bearer $HUMANIZERAI_API_KEY
Content-Type: application/json

{
  "text": "<user's text>",
  "intensity": "medium"
}

Response Format

Present results like this:

## Humanization Complete

**Score:** X → Y (improvement)
**Words Processed:** N
**Credits Remaining:** X

---
### Humanized Text

[The humanized text]

---

[Recommendation based on final score]

Credit Usage

• 1 word = 1 credit
• Detection is free
• Check credits at https://humanizerai.com/dashboard

Error Handling

Insufficient Credits

If the user doesn't have enough credits:

1. Show how many credits are needed vs available
2. Direct them to https://humanizerai.com/dashboard to top up

Invalid API Key

1. Check HUMANIZERAI_API_KEY environment variable
2. Direct to https://humanizerai.com to get a key

Rate Limit

If rate limited, suggest waiting or upgrading to Business plan.