Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

An MCP extension package for OpenAI Agents SDK

License

NotificationsYou must be signed in to change notification settings

lastmile-ai/openai-agents-mcp

 
 

Repository files navigation

This package extends theOpenAI Agents SDK to add support for Model Context Protocol (MCP) servers. With this extension, you can seamlessly use MCP servers and their tools with the OpenAI Agents SDK.

The project is built using themcp-agent library.

discordPepy Total Downloads

Features

  • Connect OpenAI Agents to MCP servers
  • Access tools from MCP servers alongside native OpenAI Agent SDK tools
  • Configure MCP servers via standard configuration files
  • Automatic tool discovery and conversion from MCP to Agent SDK format

Installation

uv add openai-agents-mcp
pip install openai-agents-mcp

Quick Start

Tip

Theexamples directory has several example applications to get started with.To run an example, clone this repo, then:

cd examplescp mcp_agent.secrets.yaml.example mcp_agent.secrets.yaml# Update API keys if neededuv run hello_world_mcp.py# Or any other example

In order to use Agents SDK with MCP, simply replace the following import:

- from agents import Agent+ from agents_mcp import Agent

With that you can instantiate an Agent withmcp_servers in addition totools (which continue to work like before).

fromagents_mcpimportAgent# Create an agent with specific MCP servers you want to use# These must be defined in your mcp_agent.config.yaml fileagent=Agent(name="MCP Agent",instructions="""You are a helpful assistant with access to both local/OpenAI tools and tools from MCP servers. Use these tools to help the user.""",# Local/OpenAI toolstools=[get_current_weather],# Specify which MCP servers to use# These must be defined in your mcp_agent configmcp_servers=["fetch","filesystem"],    )

Then define anmcp_agent.config.yaml, with the MCP server configuration:

mcp:servers:fetch:command:npxargs:["-y", "@modelcontextprotocol/server-fetch"]filesystem:command:npxargs:["-y", "@modelcontextprotocol/server-filesystem", "."]

That's it! The rest of the Agents SDK works exactly as before.

Head over to theexamples directory to see MCP servers in action with Agents SDK.

Demo

openai_agents_mcp.mov

More details and nuances below.

Using MCP servers in Agents SDK

mcp_servers property on Agent

You can specify the names of MCP servers to give an Agent access to bysetting itsmcp_servers property.

The Agent will then automatically aggregate tools from the servers, as well asanytools specified, and create a single extended list of tools. This means you can seamlesslyuse local tools, MCP servers, and other kinds of Agent SDK tools through a single unified syntax.

agent=Agent(name="MCP Assistant",instructions="You are a helpful assistant with access to MCP tools.",tools=[your_other_tools],# Regular tool use for Agent SDKmcp_servers=["fetch","filesystem"]# Names of MCP servers from your config file (see below))

MCP Configuration File

Configure MCP servers by creating anmcp_agent.config.yaml file. You can place this file in your project directory or any parent directory.

Here's an example configuration file that defines three MCP servers:

$schema:"https://raw.githubusercontent.com/lastmile-ai/mcp-agent/main/schema/mcp-agent.config.schema.json"mcp:servers:fetch:command:"uvx"args:["mcp-server-fetch"]filesystem:command:"npx"args:["-y", "@modelcontextprotocol/server-filesystem", "."]slack:command:"npx"args:["-y", "@modelcontextprotocol/server-slack"]

For servers that require sensitive information like API keys, you can:

  1. Define them directly in the config file (not recommended for production)
  2. Use a separatemcp_agent.secrets.yaml file (more secure)
  3. Set them as environment variables

Methods for Configuring MCP

This extension supports several ways to configure MCP servers:

1. Automatic Discovery (Recommended)

The simplest approach lets the SDK automatically find your configuration file if it's namedmcp_agent.config.yaml andmcp_agent.secrets.yaml:

fromagents_mcpimportAgent,RunnerContext# Create an agent that references MCP serversagent=Agent(name="MCP Assistant",instructions="You are a helpful assistant with access to MCP tools.",mcp_servers=["fetch","filesystem"]# Names of servers from your config file)result=awaitRunner.run(agent,input="Hello world",context=RunnerContext())

2. Explicit Config Path

You can explicitly specify the path to your config file:

fromagents_mcpimportRunnerContextcontext=RunnerContext(mcp_config_path="/path/to/mcp_agent.config.yaml")

3. Programmatic Configuration

You can programmatically define your MCP settings:

frommcp_agent.configimportMCPSettings,MCPServerSettingsfromagents_mcpimportRunnerContext# Define MCP config programmaticallymcp_config=MCPSettings(servers={"fetch":MCPServerSettings(command="uvx",args=["mcp-server-fetch"]        ),"filesystem":MCPServerSettings(command="npx",args=["-y","@modelcontextprotocol/server-filesystem","."]        )    })context=RunnerContext(mcp_config=mcp_config)

4. Custom Server Registry

You can create and configure your own MCP server registry:

frommcp_agent.mcp_server_registryimportServerRegistryfrommcp_agent.configimportget_settingsfromagents_mcpimportAgent# Create a custom server registrysettings=get_settings("/path/to/config.yaml")server_registry=ServerRegistry(config=settings)# Create an agent with this registryagent=Agent(name="Custom Registry Agent",instructions="You have access to custom MCP servers.",mcp_servers=["fetch","filesystem"],mcp_server_registry=server_registry# Use custom registry)

Examples

Basic Hello World

A simple example demonstrating how to create an agent that uses MCP tools:

fromagents_mcpimportAgent,RunnerContext# Create an agent with MCP serversagent=Agent(name="MCP Assistant",instructions="You are a helpful assistant with access to tools.",tools=[get_current_weather],# Local toolsmcp_servers=["fetch","filesystem"],# MCP servers)# Run the agentresult=awaitRunner.run(agent,input="What's the weather in Miami? Also, can you fetch the OpenAI website?",context=RunnerContext(),)print(result.response.value)

Seehello_world_mcp.py for the complete example.

Streaming Responses

To stream responses instead of waiting for the complete result:

result=Runner.run_streamed(# Note: No await hereagent,input="Print the first paragraph of https://openai.github.io/openai-agents-python/",context=context,)# Stream the eventsasyncforeventinresult.stream_events():ifevent.type=="raw_response_event"andisinstance(event.data,ResponseTextDeltaEvent):print(event.data.delta,end="",flush=True)

Seehello_world_mcp_streamed.py for the complete example.

Acknowledgements

This project is made possible thanks to the following projects:

License

MIT

[8]ページ先頭
©2009-2025 Movatter.jp