GPTZero

MCP server that detects AI-generated text via the GPTZero API with multilingual support and detailed confidence insights.

javascript

0

GitHub Stars

javascript

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "louis030195-gptzero-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@louis030195/gptzero-mcp"
      ],
      "env": {
        "GPTZERO_API_KEY": "your-gptzero-api-key"
      }
    }
  }
}

You can run the GPTZero MCP server to detect AI-generated text directly through the GPTZero API. It exposes a dedicated MCP endpoint that lets you analyze text, get confidence scores, and receive multilingual results for French and Spanish when enabled.

How to use

Interact with the GPTZero MCP server using a client that supports MCP. You configure a client to connect via the MCP channel, then send text to be analyzed. You will get back a predicted class (ai, human, or mixed), a confidence category, probability scores for each class, and sentence-level details that help you understand why a result was produced.

How to install

Prerequisites: you need Node.js installed on your system and access to the GPTZero API key. You will run a local or remote MCP client that points to the GPTZero MCP server.

Step 1. Install and configure the MCP client to run GPTZero MCP. Use the following configuration example to register the MCP server as a stdio adapter via the client you are using.

Step 2. Ensure your GPTZERO_API_KEY is available in the environment where you run the MCP client.

Additional configuration and examples

The GPTZero MCP server provides tools to detect AI-generated text and to list available GPTZero model versions. You can enable multilingual detection for French and Spanish with the multilingual flag.

Tools you can access include:

gptzero_detect: Detect if text was generated by AI with a detailed probability breakdown and sentence-level analysis. Multilingual detection is optional.
gptzero_model_versions: Retrieve available GPTZero model versions.

CLI and configuration examples

{
  "mcpServers": {
    "gptzero": {
      "command": "npx",
      "args": ["-y", "@louis030195/gptzero-mcp"],
      "env": {
        "GPTZERO_API_KEY": "your-gptzero-api-key"
      }
    }
  }
}

Development

To run locally, install dependencies, build, and start the MCP server with your API key.

# Install dependencies
npm install

# Build the project (if applicable)
npm run build

# Run locally with your GPTZero API key
GPTZERO_API_KEY=your-api-key npm start

Available tools

gptzero_detect

Detect if provided text was generated by AI, returning a classified result (ai, human, or mixed), confidence level, per-class probabilities, and sentence-level analysis.

gptzero_model_versions

Return a list of available GPTZero model versions that can be used by the MCP server.