OpenAI WebSearch MCP Server

This MCP server enables AI assistants to access OpenAI's web search capabilities via the Model Context Protocol. It allows assistants to augment their knowledge with real-time web information during conversations. The server is designed for integration with Claude.app and Zed editor.

One-Click Installation & Configuration

Claude

OPENAI_API_KEY=sk-xxxx uv run --with uv --with openai-websearch-mcp openai-websearch-mcp-install

This command automatically configures the server, using your OpenAI API key.

Cursor & Windsurf

Coming soon.

Manual Installation and Configuration

Ensure uvx is installed.

Claude:

1. Using uvx:

"mcpServers": { "openai-websearch-mcp": { "command": "uvx", "args": ["openai-websearch-mcp"], "env": { "OPENAI_API_KEY": "your-api-key-here" } } }

2. Using pip:

pip install openai-websearch-mcp

"mcpServers": { "openai-websearch-mcp": { "command": "python", "args": ["-m", "openai_websearch_mcp"], "env": { "OPENAI_API_KEY": "your-api-key-here" } } }

Zed:

Use similar JSON configurations within Zed's settings.json, adapting the "mcpServers" key to "context_servers".

Debugging

Use the MCP inspector:

npx @modelcontextprotocol/inspector uvx openai-websearch-mcp