UNPKG

@boseong/mcp-openai-gpt5

Version:

MCP server for OpenAI GPT-5 API with dual interface support (simple + messages)

217 lines (159 loc) 5.63 kB
# @boseong/mcp-openai-gpt5 MCP (Model Context Protocol) server for OpenAI GPT-5 API integration with dual interface support. This server provides two complementary ways to interact with OpenAI's GPT-5 models through the MCP protocol. ## Features - **Dual Interface Support**: Two complementary tools for different use cases - `gpt5_generate`: Simple text generation with prompts - `gpt5_messages`: Structured conversations with message arrays - **Flexible Configuration**: Support for reasoning effort levels (low, medium, high) - **Comprehensive Parameters**: Temperature, top-p, token limits, system instructions - **Detailed Usage Tracking**: Token consumption reporting for both prompt and completion - **Smart Environment Handling**: Multiple .env file location support for different deployment scenarios ## What Makes This Different This server differs from existing GPT-5 MCP implementations by providing: 1. **Two Distinct Tools**: Unlike single-tool servers, this provides both simple and structured interfaces 2. **Enhanced Token Reporting**: Detailed usage statistics including prompt/completion/total tokens 3. **Flexible Environment**: Supports multiple .env file locations for various deployment setups 4. **Message-First Design**: Built specifically for structured conversation workflows ## Installation ### Using NPM (Global Installation) ```bash npm install -g @boseong/mcp-openai-gpt5 ``` ### Using NPX (No Installation Required) You can run the server directly with npx without installing: ```bash npx @boseong/mcp-openai-gpt5 ``` ## Configuration ### Setting up OpenAI API Key You need an OpenAI API key to use this server. You can set it in two ways: #### Option 1: Environment Variable Set the `OPENAI_API_KEY` environment variable: ```bash export OPENAI_API_KEY=your-api-key-here ``` #### Option 2: Claude Desktop Configuration Add the API key directly in your Claude Desktop configuration file. ### Claude Code Integration To use this server with Claude Code, you can use either `gpt5` or `openai-gpt5` as the server name. Use the `mcp add` command: #### Option 1: Short Name (Recommended) ```bash # Add the MCP server using the short name "gpt5" claude mcp add gpt5 --env OPENAI_API_KEY=your-openai-api-key-here \ -- npx @boseong/mcp-openai-gpt5 ``` #### Option 2: Full Name (Legacy) ```bash # Add the MCP server using the full name "openai-gpt5" claude mcp add openai-gpt5 --env OPENAI_API_KEY=your-openai-api-key-here \ -- npx @boseong/mcp-openai-gpt5 ``` Or using the add-json command for more complex configuration: ```bash # Using short name claude mcp add-json gpt5 --scope user '{ "command": "npx", "args": ["@boseong/mcp-openai-gpt5"], "env": { "OPENAI_API_KEY": "your-openai-api-key-here" } }' # Using full name claude mcp add-json openai-gpt5 --scope user '{ "command": "npx", "args": ["@boseong/mcp-openai-gpt5"], "env": { "OPENAI_API_KEY": "your-openai-api-key-here" } }' ``` ### Claude Desktop Integration To use this server with Claude Desktop, add it to your Claude Desktop configuration: **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json` **Windows**: `%APPDATA%/Claude/claude_desktop_config.json` You can use either `gpt5` (recommended) or `openai-gpt5` as the server key: #### Option 1: Short Name Configuration (Recommended) ```json { "mcpServers": { "gpt5": { "command": "npx", "args": ["@boseong/mcp-openai-gpt5"], "env": { "OPENAI_API_KEY": "your-openai-api-key-here" } } } } ``` #### Option 2: Full Name Configuration (Legacy) ```json { "mcpServers": { "openai-gpt5": { "command": "npx", "args": ["@boseong/mcp-openai-gpt5"], "env": { "OPENAI_API_KEY": "your-openai-api-key-here" } } } } ``` If you installed globally, you can also use: ```json { "mcpServers": { "gpt5": { "command": "mcp-openai-gpt5", "env": { "OPENAI_API_KEY": "your-openai-api-key-here" } } } } ``` ## Available Tools ### gpt5_generate Simple text generation with a prompt. **Parameters:** - `input` (required): The prompt text - `model` (optional): Model variant (default: "gpt-5") - `instructions` (optional): System instructions for the model - `reasoning_effort` (optional): "low" | "medium" | "high" - `max_tokens` (optional): Maximum tokens to generate - `temperature` (optional): 0-2 range for randomness - `top_p` (optional): 0-1 range for nucleus sampling ### gpt5_messages Structured conversation with message arrays. **Parameters:** - `messages` (required): Array of conversation messages - `role`: "user" | "developer" | "assistant" - `content`: Message content - Same optional parameters as `gpt5_generate` ## Usage Examples Once configured in Claude Desktop, you can use the GPT-5 tools directly in your conversations: 1. **Simple generation**: Ask Claude to use GPT-5 to generate text 2. **Conversation mode**: Use GPT-5 with structured messages for multi-turn conversations 3. **Reasoning tasks**: Enable high reasoning effort for complex problems ## Development ### Building from Source ```bash git clone https://github.com/yourusername/mcp-server-gpt5.git cd mcp-server-gpt5 npm install npm run build ``` ### Running Locally ```bash npm start ``` ## Requirements - Node.js >= 18.0.0 - OpenAI API key with GPT-5 access ## License MIT ## Contributing Contributions are welcome! Please feel free to submit a Pull Request. ## Support For issues and questions, please visit the [GitHub repository](https://github.com/yourusername/mcp-server-gpt5).