MCP Servers

A collection of Model Context Protocol servers, templates, tools and more.

MCP server for Pangram Labs AI text detection API

Created 4/10/2026
Updated about 5 hours ago
Repository documentation and setup instructions

pangram-mcp

MCP server for Pangram Labs AI text detection. Exposes a single analyze tool that classifies text as human-written, AI-generated, or AI-assisted.

Setup

Get an API key from Pangram Labs and set it in your environment:

export PANGRAM_API_KEY="your-key-here"

Build

cargo build --release

Usage

The server communicates over stdio using the Model Context Protocol. Add it to your MCP client config:

Claude Code

claude mcp add pangram -- /path/to/pangram-mcp/target/release/pangram-mcp

Claude Desktop / generic MCP client

{
  "mcpServers": {
    "pangram": {
      "command": "/path/to/pangram-mcp/target/release/pangram-mcp",
      "env": {
        "PANGRAM_API_KEY": "your-key-here"
      }
    }
  }
}

Tool

analyze

Detect AI-generated text. Provide either raw text or a file path.

| Parameter | Type | Description | |-----------|--------|--------------------------------| | text | string | Raw text to analyze | | file | string | Path to a file to read and analyze |

Returns per-segment classification with AI scores and confidence levels.

Quick Setup
Installation guide for this server

Installation Command (package not published)

git clone https://github.com/nm-z/pangram-mcp
Manual Installation: Please check the README for detailed setup instructions and any additional dependencies required.

Cursor configuration (mcp.json)

{ "mcpServers": { "nm-z-pangram-mcp": { "command": "git", "args": [ "clone", "https://github.com/nm-z/pangram-mcp" ] } } }