GPTHuman Humanizer
@GPTHuman-ai
MCP server providing access to GPTHuman's API, the leading platform for rewriting AI-generated text into more natural, human-sounding prose, that successfully bypasses AI detectors.
Overview
What is GPTHuman Humanizer?
GPTHuman Humanizer is a Model Context Protocol (MCP) server that provides access to GPTHuman’s API for rewriting AI-generated text into more natural, human-sounding prose. It is intended for use with any MCP-compatible client such as Cursor or Claude Desktop.
How to use GPTHuman Humanizer?
Set the GPTHUMAN_API_KEY environment variable to a valid API key from GPTHuman.ai, then run npx -y @gpthuman/mcp-server. Configure the server in your client’s MCP settings (e.g., ~/.cursor/mcp.json or claude_desktop_config.json) by specifying the command and environment variable. The server exposes a single humanize_text tool.
Key features of GPTHuman Humanizer
- Single
humanize_texttool for rewriting AI text. - Supports four tones: Standard, HighSchool, College, PhD.
- Three rewrite modes: Professional, Balanced, Enhanced.
- Returns a human score and metadata with each rewrite.
- Input must be 300–2,000 characters; output includes word/credit counts.
Use cases of GPTHuman Humanizer
- Making AI-generated content read naturally to bypass detectors.
- Polishing academic or professional writing into a conversational tone.
- Adjusting the reading level of text for different audiences.
- Obtaining a human-score estimate and other analysis metadata.
FAQ from GPTHuman Humanizer
What are the runtime requirements?
Node.js version 22.0.0 or newer and a GPTHuman API key are required.
Where do I get an API key?
An API key can be obtained from the GPTHuman app dashboard at GPTHuman.ai/api.
What are the text size limits?
The input text must be at least 300 characters and at most 2,000 words.
What transport does the server use?
The server communicates over the standard I/O (stdio) transport, compatible with all MCP clients that support it.
What metadata is returned?
The tool returns a markdown summary containing the human score, similarity to the original, readability level, detected language, applied tone and mode, word and character counts, credit usage, remaining balance, and a request ID.
