How to integrate Gmail MCP with OpenAI Agents SDK

BOOK A CALLGET STARTED FOR FREE
Framework Integration Gradient
Gmail Logo
open-ai-agents-sdk Logo
divider

Introduction

This guide walks you through connecting Gmail to the OpenAI Agents SDK using the Composio tool router. By the end, you'll have a working Gmail agent that can read emails, search your inbox, draft messages, manage labels, and organize threads through natural language commands.

This guide will help you understand how to give your OpenAI Agents SDK agent real control over a Gmail account through Composio's Gmail MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

TL;DR

Here's what you'll learn:
  • Get and set up your OpenAI and Composio API keys
  • Install the necessary dependencies
  • Initialize Composio and create a Tool Router session for Gmail
  • Configure an AI agent that can use Gmail as a tool
  • Run a live chat session where you can ask the agent to perform Gmail operations

What is open-ai-agents-sdk?

The OpenAI Agents SDK is a lightweight framework for building AI agents that can use tools and maintain conversation state. It provides a simple interface for creating agents with hosted MCP tool support.

Key features include:

  • Hosted MCP Tools: Connect to external services through hosted MCP endpoints
  • SQLite Sessions: Persist conversation history across interactions
  • Simple API: Clean interface with Agent, Runner, and tool configuration
  • Streaming Support: Real-time response streaming for interactive applications

What is the Gmail MCP server, and what's possible with it?

The Gmail MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Gmail account. It provides structured and secure access to your email, so your agent can search, read, draft, organize, and even manage contacts in your mailbox—all on your behalf.

  • Advanced email search and retrieval: Effortlessly instruct your agent to fetch emails by sender, subject, label, date, or keywords, and even retrieve full message content or threads.
  • Automated drafting and sending: Have your agent create new email drafts, craft replies, add CC/BCC, include attachments, and handle threading to streamline communication.
  • Smart label and inbox organization: Let the agent create new labels, apply or remove labels from emails, and keep your inbox clutter-free by archiving or moving messages.
  • Contact and thread management: Fetch your Gmail contacts, pull entire conversation threads, or download specific attachments to make follow-ups a breeze.
  • Email and draft cleanup: Direct your agent to permanently delete emails or drafts, helping you maintain a tidy mailbox with minimal effort.

Supported Tools & Triggers

Tools
Triggers
Modify email labelsAdds and/or removes specified gmail labels for a message; ensure `message id` and all `label ids` are valid (use 'listlabels' for custom label ids).
Create email draftCreates a gmail email draft, supporting to/cc/bcc, subject, plain/html body (ensure `is html=true` for html), attachments, and threading.
Create labelCreates a new label with a unique name in the specified user's gmail account.
Delete DraftPermanently deletes a specific gmail draft using its id; ensure the draft exists and the user has necessary permissions for the given `user id`.
Delete messagePermanently deletes a specific email message by its id from a gmail mailbox; for `user id`, use 'me' for the authenticated user or an email address to which the authenticated user has delegated access.
Fetch emailsFetches a list of email messages from a gmail account, supporting filtering, pagination, and optional full content retrieval.
Fetch message by message IDFetches a specific email message by its id, provided the `message id` exists and is accessible to the authenticated `user id`.
Fetch Message by Thread IDRetrieves messages from a gmail thread using its `thread id`, where the thread must be accessible by the specified `user id`.
Get Gmail attachmentRetrieves a specific attachment by id from a message in a user's gmail mailbox, requiring valid message and attachment ids.
Get contactsFetches contacts (connections) for the authenticated google account, allowing selection of specific data fields and pagination.
Get PeopleRetrieves either a specific person's details (using `resource name`) or lists 'other contacts' (if `other contacts` is true), with `person fields` specifying the data to return.
Get ProfileRetrieves key gmail profile information (email address, message/thread totals, history id) for a user.
List draftsRetrieves a paginated list of email drafts from a user's gmail account.
List Gmail labelsRetrieves a list of all system and user-created labels for the specified gmail account.
List threadsRetrieves a list of email threads from a gmail account, identified by `user id` (email address or 'me'), supporting filtering and pagination.
Modify thread labelsAdds or removes specified existing label ids from a gmail thread, affecting all its messages; ensure the thread id is valid.
Move to TrashMoves an existing, non-deleted email message to the trash for the specified user.
Patch LabelPatches the specified label.
Remove labelPermanently deletes a specific, existing user-created gmail label by its id for a user; cannot delete system labels.
Reply to email threadSends a reply within a specific gmail thread using the original thread's subject, requiring a valid `thread id` and correctly formatted email addresses.
Search PeopleSearches contacts by matching the query against names, nicknames, emails, phone numbers, and organizations, optionally including 'other contacts'.
Send DraftSends the specified, existing draft to the recipients in the to, cc, and bcc headers.
Send EmailSends an email via gmail api using the authenticated user's google profile display name, requiring `is html=true` if the body contains html and valid `s3key`, `mimetype`, `name` for any attachment.

What is the Composio tool router, and how does it fit here?

What is Tool Router?

Composio's Tool Router helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Tool Router

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Tool Router works

The Tool Router follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before starting, make sure you have:
  • Composio API Key and OpenAI API Key
  • Primary know-how of OpenAI Agents SDK
  • A live Gmail project
  • Some knowledge of Python or Typescript

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key

Install dependencies

pip install composio_openai_agents openai-agents python-dotenv

Install the Composio SDK and the OpenAI Agents SDK.

Set up environment variables

bash
OPENAI_API_KEY=sk-...your-api-key
COMPOSIO_API_KEY=your-api-key
USER_ID=composio_user@gmail.com

Create a .env file and add your OpenAI and Composio API keys.

Import dependencies

import asyncio
import os
from dotenv import load_dotenv

from composio import Composio
from composio_openai_agents import OpenAIAgentsProvider
from agents import Agent, Runner, HostedMCPTool, SQLiteSession
What's happening:
  • You're importing all necessary libraries.
  • The Composio and OpenAIAgentsProvider classes are imported to connect your OpenAI agent to Composio tools like Gmail.

Set up the Composio instance

load_dotenv()

api_key = os.getenv("COMPOSIO_API_KEY")
user_id = os.getenv("USER_ID")

if not api_key:
    raise RuntimeError("COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key")

# Initialize Composio
composio = Composio(api_key=api_key, provider=OpenAIAgentsProvider())
What's happening:
  • load_dotenv() loads your .env file so OPENAI_API_KEY and COMPOSIO_API_KEY are available as environment variables.
  • Creating a Composio instance using the API Key and OpenAIAgentsProvider class.

Create a Tool Router session

# Create a Gmail Tool Router session
session = composio.create(
    user_id=user_id,
    toolkits=["gmail"]
)

mcp_url = session.mcp.url

What is happening:

  • You give the Tool Router the user id and the toolkits you want available. Here, it is only gmail.
  • The router checks the user's Gmail connection and prepares the MCP endpoint.
  • The returned session.mcp.url is the MCP URL that your agent will use to access Gmail.
  • This approach keeps things lightweight and lets the agent request Gmail tools only when needed during the conversation.

Configure the agent

# Configure agent with MCP tool
agent = Agent(
    name="Assistant",
    model="gpt-5",
    instructions=(
        "You are a helpful assistant that can access Gmail. "
        "Help users perform Gmail operations through natural language."
    ),
    tools=[
        HostedMCPTool(
            tool_config={
                "type": "mcp",
                "server_label": "tool_router",
                "server_url": mcp_url,
                "headers": {"x-api-key": api_key},
                "require_approval": "never",
            }
        )
    ],
)
What's happening:
  • We're creating an Agent instance with a name, model (gpt-5), and clear instructions about its purpose.
  • The agent's instructions tell it that it can access Gmail and help with queries, inserts, updates, authentication, and fetching database information.
  • The tools array includes a HostedMCPTool that connects to the MCP server URL we created earlier.
  • The headers dict includes the Composio API key for secure authentication with the MCP server.
  • require_approval: 'never' means the agent can execute Gmail operations without asking for permission each time, making interactions smoother.

Start chat loop and handle conversation

print("\nComposio Tool Router session created.")

chat_session = SQLiteSession("conversation_openai_toolrouter")

print("\nChat started. Type your requests below.")
print("Commands: 'exit', 'quit', or 'q' to end\n")

async def main():
    try:
        result = await Runner.run(
            agent,
            "What can you help me with?",
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")
    except Exception as e:
        print(f"Error: {e}\n")

    while True:
        user_input = input("You: ").strip()
        if user_input.lower() in {"exit", "quit", "q"}:
            print("Goodbye!")
            break

        result = await Runner.run(
            agent,
            user_input,
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")

asyncio.run(main())
What's happening:
  • The program prints a session URL that you visit to authorize Gmail.
  • After authorization, the chat begins.
  • Each message you type is processed by the agent using Runner.run().
  • The responses are printed to the console, and conversations are saved locally using SQLite.
  • Typing exit, quit, or q cleanly ends the chat.

Complete Code

Here's the complete code to get you started with Gmail and open-ai-agents-sdk:

import asyncio
import os
from dotenv import load_dotenv

from composio import Composio
from composio_openai_agents import OpenAIAgentsProvider
from agents import Agent, Runner, HostedMCPTool, SQLiteSession

load_dotenv()

api_key = os.getenv("COMPOSIO_API_KEY")
user_id = os.getenv("USER_ID")

if not api_key:
    raise RuntimeError("COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key")

# Initialize Composio
composio = Composio(api_key=api_key, provider=OpenAIAgentsProvider())

# Create Tool Router session
session = composio.create(
    user_id=user_id,
    toolkits=["gmail"]
)
mcp_url = session.mcp.url

# Configure agent with MCP tool
agent = Agent(
    name="Assistant",
    model="gpt-5",
    instructions=(
        "You are a helpful assistant that can access Gmail. "
        "Help users perform Gmail operations through natural language."
    ),
    tools=[
        HostedMCPTool(
            tool_config={
                "type": "mcp",
                "server_label": "tool_router",
                "server_url": mcp_url,
                "headers": {"x-api-key": api_key},
                "require_approval": "never",
            }
        )
    ],
)

print("\nComposio Tool Router session created.")

chat_session = SQLiteSession("conversation_openai_toolrouter")

print("\nChat started. Type your requests below.")
print("Commands: 'exit', 'quit', or 'q' to end\n")

async def main():
    try:
        result = await Runner.run(
            agent,
            "What can you help me with?",
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")
    except Exception as e:
        print(f"Error: {e}\n")

    while True:
        user_input = input("You: ").strip()
        if user_input.lower() in {"exit", "quit", "q"}:
            print("Goodbye!")
            break

        result = await Runner.run(
            agent,
            user_input,
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")

asyncio.run(main())

Conclusion

This was a starter code for integrating Gmail MCP with OpenAI Agents SDK to build a functional AI agent that can interact with Gmail.

Key features:

  • Hosted MCP tool integration through Composio's Tool Router
  • SQLite session persistence for conversation history
  • Simple async chat loop for interactive testing
You can extend this by adding more toolkits, implementing custom business logic, or building a web interface around the agent.

How to build Gmail MCP Agent with another framework

Claude Agent SDK

Claude Agent SDK

Use Gmail MCP with Claude Agent SDK

Google ADK

Google ADK

Use Gmail MCP with Google ADK

LangChain

LangChain

Use Gmail MCP with LangChain

Vercel AI SDK

Vercel AI SDK

Use Gmail MCP with Vercel AI SDK

Mastra AI

Mastra AI

Use Gmail MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Gmail MCP with LlamaIndex

CrewAI

CrewAI

Use Gmail MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Gmail MCP?

With a standalone Gmail MCP server, the agents and LLMs can only access a fixed set of Gmail tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Gmail and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with OpenAI Agents SDK?

Yes, you can. OpenAI Agents SDK fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Gmail tools.

Can I manage the permissions and scopes for Gmail while using Tool Router?

Yes, absolutely. You can configure which Gmail scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Gmail data and credentials are handled as safely as possible.

Used by agents from

Context
ASU
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
ASU
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
ASU
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Book a call↗Get started for free