OpenAI MCP Server

Name: OpenAI MCP Server
Author: akiojin

@akiojin

代码仓库

关于 OpenAI MCP Server

暂无概览

基本信息

分类

AI 与智能体

许可证

MIT license

运行时

node

传输方式

stdio

发布者

akiojin

配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块，请前往代码仓库查看安装说明。

代码仓库

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is OpenAI MCP Server?

OpenAI MCP Server is a Model Context Protocol (MCP) server that enables AI assistants to interact with OpenAI’s ChatGPT API. It provides tools for generating text completions and listing available models, targeting developers who integrate AI into MCP-compatible clients.

How to use OpenAI MCP Server?

Prerequisites are Node.js 18+ and an OpenAI API key. Clone the repository, run npm install and npm run build, then create a .env file with OPENAI_API_KEY. Add the server to your MCP client configuration using the command node with args pointing to the built dist/index.js. The server exposes two tools: chat_completion and list_models.

Key features of OpenAI MCP Server

Chat completion with multiple OpenAI models (GPT-4.1, GPT-4.1-mini, GPT-4.1-nano, GPT-4o, o1, o1-pro, o3, o4-mini)
List all available OpenAI models
Configurable parameters: temperature (0–2), max tokens (default 1000), and model selection
Built on the Model Context Protocol for seamless AI assistant integration

Use cases of OpenAI MCP Server

Integrate ChatGPT-powered text generation into MCP-compatible AI assistants
Dynamically select and query OpenAI models based on task requirements
Prototype and deploy conversational agents with configurable generation settings
Retrieve model lists for tool selection or resource planning

FAQ from OpenAI MCP Server

What models are supported?

The server supports GPT-4.1, GPT-4.1-mini, GPT-4.1-nano, GPT-4o, o1, o1-pro, o3, and o4-mini, with GPT-4.1 as the default.

How do I configure the API key?

Copy .env.example to .env in the project root, then set OPENAI_API_KEY=your_api_key_here.

What are the system requirements?

Node.js 18 or higher and an active OpenAI API key are required.

How do I use the chat_completion tool?

Call chat_completion with required messages (array of role/content objects). Optional parameters include model, temperature (0–2, default 0.7), and max_tokens (default 1000).