Openai Agents

🚀 OpenAI Agents MCP Server

A server that exposes OpenAI agents via the MCP protocol.

🚀 Quick Start

This is a server that exposes OpenAI agents through the MCP protocol, providing both single OpenAI agents and a multi - agent coordinator, which can be accessed by any MCP client.

✨ Features

This server provides both a single OpenAI agent and a multi - agent coordinator, both using OpenAI's agents SDK:

Dedicated Single Agents

• Web Search Agent: A dedicated agent for real - time information queries.
• File Search Agent: An agent for searching and analyzing files in the OpenAI vector store.
• Computer Operation Agent: A dedicated agent for securely performing computer operations.

Multi - Agent Coordinator

• Coordinator Agent: A powerful agent capable of coordinating each dedicated agent and selecting the most suitable agent (or multiple agents) to complete a task.

Each agent can be accessed via the MCP protocol and can be used by any MCP client, including the Claude desktop application.

📦 Installation

Prerequisites

• Python 3.11 or higher
• uv package manager (recommended)
• OpenAI API key

Installation via Smithery

To automatically install openai-agents-mcp-server via Smithery:

npx -y @smithery/cli install @lroolle/openai-agents-mcp-server --client claude

Claude Desktop Version

Add the following content to config.json to configure the Claude desktop version:

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

🔧 Technical Details

Tool Requirements

• WebSearchTool: No parameters required, but can accept a context location.
• FileSearchTool: Requires vector_store_ids (IDs from the OpenAI vector store).
• ComputerTool: Requires a complete implementation of the AsyncComputer interface (currently a mock version).

Customization

You can customize this server in the following ways:

1. Implement a complete AsyncComputer interface to enable real computer interaction.
2. Add dedicated agents for other OpenAI tools.
3. Enhance the capabilities of the coordinator agent to handle more complex processes.

📚 Documentation

The server can be configured using environment variables:

• OPENAI_API_KEY: Your OpenAI API key (required).
• MCP_TRANSPORT: The transport protocol to use, defaults to "stdio", optional "sse".

💻 Usage Examples

Setting up the Development Environment

# Clone the repository
git clone https://github.com/lroolle/openai-agents-mcp-server.git
cd openai-agents-mcp-server

# Create a virtual environment
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
uv sync --dev

Testing with the MCP Checker

You can test using the MCP checker with the following steps:

# In one terminal window, run the server with SSE transport
export OPENAI_API_KEY=your-api-key
export MCP_TRANSPORT=sse

uv run mcp dev src/agents_mcp_server/server.py

Then open http://localhost:5173 in your browser.

📄 License

MIT