o3-search MCP

o3-search-mcp

An MCP server that brings the power of OpenAI's o3 model to your AI agents, enabling them to perform intelligent web searches with natural language queries.

Usage Examples

Once installed, your AI agent can use the o3-search tool to perform web searches. For instance, try giving instructions like this:

🐛 When debugging gets tough

Use "ask o3" to find solutions from GitHub issues and Stack Overflow:

"I'm getting a 'Module not found' error in Next.js 14. Ask o3 for recent solutions"
"Debug this WebSocket connection issue. Try asking o3 for help"

🧩 When tackling complex tasks

Add "consult o3 if you get stuck" to your requests:

"Implement a distributed caching system with Redis. If you encounter difficulties, consult o3"
"Create a real-time collaborative editor. Ask o3 for help if you get stuck"

📚 For latest library info and migration guides

Stay up-to-date with "ask o3":

"How do I migrate from React Router v5 to v6? Ask o3 for the latest migration guide"
"What's the current best practice for state management in React? Ask o3 for recent recommendations"

When you make a request to your coding agent, it can autonomously consult o3 by using the MCP interface to exchange natural language queries and responses. Your agent and o3 work together in real time to help you solve problems.

Installation

Using npx (Recommended)

Claude Code:

$ claude mcp add o3 -s user \
	-e OPENAI_API_KEY=your-api-key \
	-e SEARCH_CONTEXT_SIZE=medium \
	-e REASONING_EFFORT=medium \
	-e OPENAI_API_TIMEOUT=60000 \
	-e OPENAI_MAX_RETRIES=3 \
	-- npx o3-search-mcp

json:

{
  "mcpServers": {
    "o3-search": {
      "command": "npx",
      "args": ["o3-search-mcp"],
      "env": {
        "OPENAI_API_KEY": "your-api-key",
        // Optional: low, medium, high (default: medium)
        "SEARCH_CONTEXT_SIZE": "medium",
        "REASONING_EFFORT": "medium",
        // Optional: API timeout in milliseconds (default: 60000)
        "OPENAI_API_TIMEOUT": "60000",
        // Optional: Maximum retry attempts (default: 3)
        "OPENAI_MAX_RETRIES": "3"
      }
    }
  }
}

Local Development Setup

If you want to download and run the code locally:

# setup
git clone git@github.com:yoshiko-pg/o3-search-mcp.git
cd o3-search-mcp
pnpm install
pnpm build

Claude Code:

$ claude mcp add o3 -s user \
	-e OPENAI_API_KEY=your-api-key \
	-e SEARCH_CONTEXT_SIZE=medium \
	-e REASONING_EFFORT=medium \
	-e OPENAI_API_TIMEOUT=60000 \
	-e OPENAI_MAX_RETRIES=3 \
	-- node /path/to/o3-search-mcp/build/index.js

json:

{
  "mcpServers": {
    "o3-search": {
      "command": "node",
      "args": ["/path/to/o3-search-mcp/build/index.js"],
      "env": {
        "OPENAI_API_KEY": "your-api-key",
        // Optional: low, medium, high (default: medium)
        "SEARCH_CONTEXT_SIZE": "medium",
        "REASONING_EFFORT": "medium",
        // Optional: API timeout in milliseconds (default: 60000)
        "OPENAI_API_TIMEOUT": "60000",
        // Optional: Maximum retry attempts (default: 3)
        "OPENAI_MAX_RETRIES": "3"
      }
    }
  }
}

Configuration

Environment Variables

OPENAI_API_KEY (required): Your OpenAI API key
SEARCH_CONTEXT_SIZE (optional): Controls the search context size
- Values: low, medium, high
- Default: medium
REASONING_EFFORT (optional): Controls the reasoning effort level
- Values: low, medium, high
- Default: medium
OPENAI_API_TIMEOUT (optional): API request timeout in milliseconds
- Default: 60000 (60 seconds)
- Example: 120000 for 2 minutes
OPENAI_MAX_RETRIES (optional): Maximum number of retry attempts for failed requests
- Default: 3
- The SDK automatically retries on rate limits (429), server errors (5xx), and connection errors

Install Server

HTTP connection URL

security – no known vulnerabilities

license - permissive license

quality - confirmed to work

How are these scores calculated?

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

Tools

o3-search

An MCP server that enables web search capabilities using OpenAI's o3 model, allowing AI assistants to perform text-based web searches and return AI-powered results.

Related MCP Servers

Google Search MCP Server
mixelpixx
A
security
F
license
A
quality
An MCP (Model Context Protocol) server that provides Google search capabilities and webpage content analysis tools. This server enables AI models to perform Google searches and analyze webpage content programmatically.
Last updated -
3
167
52
TypeScript
ReActMCP Web Search
mshojaei77
-
security
F
license
-
quality
An MCP server that integrates real-time web search capabilities into AI assistants using the Exa API, providing both basic and advanced search functionality with formatted markdown results.
Last updated -
119
Python
4o-mini-search-mcp
rebers
-
security
F
license
-
quality
An MCP server that enables AI models to search the web using OpenAI's 4o-mini Search model, allowing access to up-to-date information for just a few cents per search.
Last updated -
1
JavaScript
Docs-MCP
herring101
A
security
A
license
A
quality
An MCP server that allows users to efficiently search and reference user-configured documents through document listing, grep searching, semantic searching with OpenAI Embeddings, and full document retrieval.
Last updated -
4
3
Python
MIT License

View all related MCP servers

o3-search MCP

o3-search-mcp

Usage Examples

🐛 When debugging gets tough

🧩 When tackling complex tasks

📚 For latest library info and migration guides

Installation

Using npx (Recommended)

Local Development Setup

Configuration

Environment Variables

Tools

Related MCP Servers

Google Search MCP Server

ReActMCP Web Search

4o-mini-search-mcp

Docs-MCP

New MCP Servers

MCP directory API