Chatgpt MCP

🚀 Using ChatGPT - MCP Server (Standard Input/Output)

This is a standard input/output server based on the Model Context Protocol (MCP). The server forwards prompts to OpenAI's ChatGPT (GPT-4o) and is designed to run within LangGraph assistants. It can achieve advanced summarization, analysis, and reasoning by accessing external large language models (LLMs).

🚀 Quick Start

✨ Features

This server exposes one tool:

{
  "name": "ask_chatgpt",
  "description": "Send the provided text ('content') to the external ChatGPT (gpt-4o) model for advanced reasoning or summarization.",
  "parameters": {
    "type": "object",
    "properties": {
      "content": {
        "type": "string",
        "description": "The text to be analyzed, summarized, compared, or reasoned about."
      }
    },
    "required": ["content"]
  }
}

Use this tool when your assistant needs to:

• Summarize long documents
• Analyze configuration files
• Compare options
• Perform advanced natural language reasoning

📦 Installation

Docker Usage

Build and run the container:

docker build -t ask-chatgpt-mcp .
docker run -e OPENAI_API_KEY=your-openai-key -i ask-chatgpt-mcp

💻 Usage Examples

Manual Testing

Test the server with a single request:

echo '{"method":"tools/call","params":{"name":"ask_chatgpt","arguments":{"content":"Summarize this configuration..."}}}' | \
  OPENAI_API_KEY=your-openai-key python3 server.py --oneshot

LangGraph Integration

To connect this MCP server to your LangGraph pipeline, configure it like this:

("chatgpt-mcp", ["python3", "server.py", "--oneshot"], "tools/discover", "tools/call")

📚 Documentation

MCP Server Configuration Example

Here is an example of configuring the server using the mcpServers JSON:

{
  "mcpServers": {
    "chatgpt": {
      "command": "python3",
      "args": [
        "server.py",
        "--oneshot"
      ],
      "env": {
        "OPENAI_API_KEY": "<YOUR_OPENAI_API_KEY>"
      }
    }
  }
}

Environment Setup

Create a .env file (automatically loaded via python-dotenv) or manually export the key:

OPENAI_API_KEY=your-openai-key

Or:

export OPENAI_API_KEY=your-openai-key

🔧 Technical Details

• Dependencies are installed during the Docker build:
  • openai
  • requests
  • python-dotenv

📄 Project Structure

.
├── Dockerfile        # Docker build for the MCP server
├── server.py         # Main stdio server implementation
└── README.md         # You're reading it!

⚠️ Important Note

• Never commit the .env file or API keys.
• Store secrets in secure environment variables or secret managers.