OpenAI MCP Server

Official OpenAI MCP server for integrating GPT models, DALL-E, and Whisper into agentic workflows. Provides tools for chat completion, image generation, audio transcription, and embeddings through the Model Context Protocol. Supports function calling, structured outputs, and streaming responses.

{
  "mcpServers": {
    "openai": {
      "command": "npx",
      "args": [
        "-y",
        "@openai/mcp-server"
      ],
      "env": {
        "OPENAI_API_KEY": "<YOUR_OPENAI_API_KEY>"
      }
    }
  }
}

{
  "type": "object",
  "required": [
    "model",
    "messages"
  ],
  "properties": {
    "model": {
      "type": "string",
      "description": "Model ID (e.g., gpt-5, gpt-5-4-mini)"
    },
    "messages": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "role": {
            "type": "string"
          },
          "content": {
            "type": "string"
          }
        }
      },
      "description": "Conversation messages"
    },
    "max_tokens": {
      "type": "integer",
      "description": "Maximum tokens in response"
    },
    "temperature": {
      "type": "number",
      "description": "Sampling temperature (0-2)"
    }
  }
}

{
  "type": "object",
  "required": [
    "prompt"
  ],
  "properties": {
    "size": {
      "type": "string",
      "description": "Image size (1024x1024, 1792x1024, 1024x1792)"
    },
    "prompt": {
      "type": "string",
      "description": "Text description of the desired image"
    },
    "quality": {
      "type": "string",
      "description": "Quality level (standard, hd)"
    }
  }
}

{
  "type": "object",
  "required": [
    "input",
    "model"
  ],
  "properties": {
    "input": {
      "type": "string",
      "description": "Text to embed"
    },
    "model": {
      "type": "string",
      "description": "Embedding model ID"
    }
  }
}