OpenAI WebSearch MCP Server

@ConechoAI

Basic information

About OpenAI WebSearch MCP Server

openai websearch tool as mcp server

OpenAI WebSearch MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "openai-websearch-mcp": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "uv",
        "--with",
        "openai-websearch-mcp",
        "openai-websearch-mcp-install"
      ],
      "env": {
        "OPENAI_API_KEY": "sk-xxxx"
      }
    }
  }
}

OpenAI WebSearch MCP Server tools

Call openai websearch as tool.

Overview

What is OpenAI WebSearch MCP Server?

This MCP server provides access to OpenAI's web search functionality through the Model Context Protocol. It enables AI assistants to search the web during conversations, retrieving up-to-date information beyond the assistant's training data. Designed for use with Claude.app and Zed editor.

How to use OpenAI WebSearch MCP Server?

Install via uvx openai-websearch-mcp or pip install openai-websearch-mcp. Set the OPENAI_API_KEY environment variable with your API key from platform.openai.com. Configure the server in your Claude settings or Zed settings.json by adding the appropriate command and arguments. Use the web_search tool, providing required arguments type (string) and search_context_size (low/medium/high), with an optional user_location object.

Key features of OpenAI WebSearch MCP Server

One-click installation for Claude with auto-configuration.
Web search tool with configurable search context size.
Optional user location data (city, country, region, timezone).
Supports both uvx and pip installation methods.
Integrates with Claude.app and Zed editor.
Debuggable via MCP inspector.

Use cases of OpenAI WebSearch MCP Server

AI assistants retrieving current news or real‑time data during a conversation.
Answering queries about recent events not covered in the training data.
Providing location‑specific search results by passing user location.
Enhancing chatbot responses with live web information in Claude or Zed.

Frequently asked questions

What does this server do?

It exposes OpenAI’s web search as an MCP tool, allowing AI assistants to perform web searches and return results within a conversation.

How do I install it?

You can install via `uvx openai-websearch-mcp` (recommended) or `pip install openai-websearch-mcp`, then configure your client with the required command and your `OPENAI_API_KEY`.

What are the dependencies?

You need `uvx` (or `pip` and Python) and a valid OpenAI API key. No other runtime dependencies are mentioned.

Which tools are available?

The server provides one tool: `web_search`. It requires `type` (string) and `search_context_size` (low/medium/high), and accepts an optional `user_location` object with city, country, region, and timezone.

Does it support editor integrations besides Claude?

Yes, the README includes configuration examples for Zed editor. Cursor and Windsurf support are listed as “Coming soon”.

Comments

More AI & Agents MCP servers

Poly-Glot AI Workspace

hmoses

1,000+ fully localized prompt templates across 40+ categories. Compare Mode — send the same prompt to up to 9 AI providers (ChatGPT, Claude, Gemini, Grok, Groq, Copilot, Mistral, Perplexity, DuckDuckGo AI) and compare an

Sphere Marketplace

dtajitdinov-aris

Sphere is an exchange where AI agents list services and post requests for other agents. Registration is one request, free, with no human approval. The first order is free: sphere_receipt_trial, one per agent per 24 hours

q-ring

I4cTime

Quantum-inspired keyring for AI coding agents. Secure secrets with superposition, entanglement, tunneling, and teleportation — CLI + 44 MCP tools.

HeyLead: LinkedIn Outreach & Lead Generation MCP Server

D4umak

HeyLead: AI agent for LinkedIn outreach. Finds the right people, writes in your voice, follows up and handles replies, from Claude Code, Cursor, any MCP client or a web dashboard.

Quest: Hire a Human

Quest (hireahuman.quest)

Hire a verified local human for small real-world tasks your AI agent can't do: errands, queues, pickups, in-person checks and on-site photos. Your agent posts a quest, local humans send offers, and your user picks one an