MCP Openai Server

🚀 OpenAI MCP Server

An implementation of the Model Context Protocol (MCP) server for the OpenAI API. This server provides a standardized interface for applications like Augment to interact with OpenAI's Large Language Models (LLMs) using the official MCP SDK.

🚀 Quick Start

This OpenAI MCP server offers a standardized way to interact with OpenAI's LLMs. It simplifies the process of integrating applications with these models, ensuring compatibility and security.

✨ Features

• 🔄 Utilize the official MCP SDK to ensure compatibility.
• Secure API key management.
• 📊 Support for conversation completion.
• 📋 Provide a list of available models.
• 🧠 Generate text embeddings.
• ⚠️ Implement proper error handling and logging.

📦 Installation

Prerequisites

• Node.js (v16 or higher)
• npm or yarn
• OpenAI API key

Steps

1. Clone this repository or copy the files to your project directory.
2. Install the dependencies:

npm install

3. Create a .env file based on the provided .env.example:

cp .env.example .env

4. Add your OpenAI API key to the .env file:

OPENAI_API_KEY=your_openai_api_key_here

💻 Usage Examples

Basic Usage

To start the server, use the following command:

npm start

For automatic restart on file changes during development:

npm run dev

Advanced Usage

Integrating with Augment

To use this MCP server with Augment, add the following to Augment's settings.json file:

"augment.advanced": {
    "mcpServers": [
        {
            "name": "openai-mcp",
            "command": "node",
            "args": ["/path/to/openai-mcp-server.js"],
            "env": {
                "OPENAI_API_KEY": "your_openai_api_key_here",
                "DEBUG": "true"
            }
        }
    ]
}

Replace /path/to/openai-mcp-server.js with the actual path to the server file.

📚 Documentation

Available Tools

This MCP server provides the following tools to Augment:

1. List Models

List all available OpenAI models.

2. Conversation Completion

Use OpenAI's conversation completion API to generate responses. Supported parameters include:

• model: The model to use (e.g., gpt-3.5-turbo, gpt-4).
• messages: An array of conversation messages.
• temperature: Control randomness (between 0 - 1).
• max_tokens: The maximum number of tokens to generate.

3. Create Embeddings

Use OpenAI's embedding API to generate embeddings for text. Supported parameters include:

• model: The model to use (e.g., text-embedding-ada-002).
• input: The text to embed (a string or an array of strings).

Testing

An example client for testing this MCP server is provided in client-example.js. Run:

node client-example.js

Understanding the Code

The server consists of several key components:

1. MCP SDK Integration: Use the official MCP SDK for standardized communication.
2. Tool Handlers: Implement handlers for each supported tool.
3. OpenAI API Integration: Communicate with OpenAI's API.
4. Error Handling: Provide consistent error responses.
5. Logging: Log operations for debugging.

Potential Improvements

See mcp-server-improvement-suggestions.md for a list of potential suggestions to improve this implementation.

📄 License

MIT