API Reference

HumanizerAI API

Integrate AI detection and humanization into your applications with our simple REST API.

Quick Start

Get an API Key

Click your profile picture → Settings → API Keys. Requires Pro or Business plan.

Add Authorization

Include your API key in the Authorization header as a Bearer token.

Make Requests

Call the detect or humanize endpoints with your text content.

Authentication

All API requests require authentication using an API key. Include your key in the Authorization header as a Bearer token:

Authorization: Bearer hum_your_api_key_here

Keep your API key secure

Never expose your API key in client-side code or public repositories. Use environment variables.

Base URL

https://humanizerai.com/api/v1

POST

/detect

Analyze text to detect if it was written by AI. Returns a score from 0-100 (higher = more likely AI) along with detailed metrics.

Request Body

Parameter	Type	Required	Description
text	string		The text to analyze (max 10,000 words)

Example Request

curl -X POST https://humanizerai.com/api/v1/detect \
  -H "Authorization: Bearer hum_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Your text to analyze for AI detection..."
  }'

Response

{
  "score": 85,
  "metrics": {
    "perplexity": 42.5,
    "burstiness": 0.35,
    "readability": 65.2,
    "satPercent": 12.5,
    "simplicity": 78.3,
    "ngramScore": 0.15,
    "averageSentenceLength": 18.4
  },
  "verdict": "likely_ai",
  "wordsProcessed": 250
}

POST

/humanize

Transform AI-generated text into natural, human-like content that bypasses AI detectors. Uses credits based on word count.

Request Body

Parameter	Type	Required	Description
text	string		The text to humanize (max 10,000 words)
intensity	string	Optional	`light`, `medium` (default), or `aggressive` (bypass)

Intensity Levels

Light

light

Subtle changes, preserves your style. Best for text with low AI scores.

Medium

medium(default)

Balanced rewrites with human touches. Best for most use cases.

Bypass

aggressive

Maximum detector bypass. Text becomes shorter and simpler.

Example Request

curl -X POST https://humanizerai.com/api/v1/humanize \
  -H "Authorization: Bearer hum_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Your AI-generated text to humanize...",
    "intensity": "medium"
  }'

Response

{
  "humanizedText": "The rewritten text that sounds human...",
  "score": {
    "before": 85,
    "after": 15
  },
  "wordsProcessed": 250,
  "credits": {
    "subscriptionRemaining": 49750,
    "topUpRemaining": 0,
    "totalRemaining": 49750
  }
}

GET

/credits

Check your current credit balance. Returns subscription credits, top-up credits, and total available.

Example Request

curl https://humanizerai.com/api/v1/credits \
  -H "Authorization: Bearer hum_your_api_key"

Response

{
  "credits": {
    "subscription": 45000,
    "topUp": 5000,
    "total": 50000
  },
  "plan": "pro",
  "billingCycleEnd": "2025-02-15T00:00:00.000Z"
}

Error Codes

Status	Message	Description
400	Text is required	Missing or empty text parameter
400	Text exceeds maximum	Text exceeds 10,000 word limit
401	API key required	Missing Authorization header
401	Invalid or revoked API key	API key not found or has been revoked
403	API access requires Pro/Business	User's plan doesn't include API access
403	insufficient_credits	Not enough credits for humanization
500	Server error	Internal server error

Rate Limits

Plan	API Keys	Rate Limit	Max Words/Request
Pro	Up to 3	60 requests/minute	10,000
Business	Up to 10	300 requests/minute	10,000

MCP Server

The MCP (Model Context Protocol) server gives Claude Desktop direct access to AI detection and humanization. Ask Claude to check or humanize text without leaving your conversation.

Install the MCP server

npm install -g @humanizerai/mcp-server

Open Claude Desktop config

Open the config file at:

# macOS
~/Library/Application Support/Claude/claude_desktop_config.json

# Windows
%APPDATA%\Claude\claude_desktop_config.json

Add HumanizerAI to your config

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "humanizerai": {
      "command": "humanizerai-mcp",
      "env": {
        "HUMANIZERAI_API_KEY": "hum_your_api_key"
      }
    }
  }
}

Restart Claude Desktop

Quit and reopen Claude Desktop. You can now ask Claude to detect or humanize text!

Available Tools

detect_ai

Check if text is AI-generated

humanize

Transform AI text to human-like

check_credits

View remaining credit balance

Agent Skills

View on GitHub

Agent skills let you detect and humanize AI text directly from your terminal with simple slash commands. Works with Claude Code, Cursor, Windsurf, and any agent supporting the Agent Skills format.

Install the skill

Option 1: /learn command (Recommended)

/learn @humanizerai/humanize

Option 2: Claude Code Plugin Marketplace

# In Claude Code, run:
/plugin marketplace add humanizerai/agent-skills
/plugin install detect-ai@humanizerai
/plugin install humanize@humanizerai

Option 3: Clone from GitHub

git clone https://github.com/humanizerai/agent-skills.git .claude/skills/humanizerai

Set your API key

export HUMANIZERAI_API_KEY="hum_your_api_key"

Use the skills

Run these commands in your agent session:

Available Commands

/detect-ai [text]

Analyze text and get AI detection score with detailed metrics

/humanize [text]

Transform AI text to bypass detectors (uses medium intensity)

/humanize --intensity aggressive [text]

Maximum bypass mode for stubborn high-AI content

Compatible Agents

AMP

And any agent supporting the Agent Skills open format.

HumanizerAI API

Quick Start

Get an API Key

Add Authorization

Make Requests

Authentication

Base URL

/detect

Request Body

Example Request

Response

/humanize

Request Body

Intensity Levels

Light

Medium

Bypass

Example Request

Response

/credits

Example Request

Response

Error Codes

Rate Limits

MCP Server

Install the MCP server

Open Claude Desktop config

Add HumanizerAI to your config

Restart Claude Desktop

Available Tools

Agent Skills

Install the skill

Set your API key

Use the skills

Available Commands

Compatible Agents

Ready to Get Started?