How to integrate Google Docs MCP with OpenAI Agents SDK

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

Introduction

This guide walks you through connecting Google Docs to the OpenAI Agents SDK using the Composio tool router. By the end, you'll have a working Google Docs agent that can create a new meeting notes document, copy last week's project summary template, add bullet points to the action items section, delete the footer from my current draft through natural language commands.

This guide will help you understand how to give your OpenAI Agents SDK agent real control over a Google Docs account through Composio's Google Docs 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 Google Docs
  • Configure an AI agent that can use Google Docs as a tool
  • Run a live chat session where you can ask the agent to perform Google Docs 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 Google Docs MCP server, and what's possible with it?

The Google Docs MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Google Docs account. It provides structured and secure access to your documents, so your agent can create, copy, edit, and organize Google Docs on your behalf.

  • Automated document creation and duplication: Let your agent generate new Google Docs from scratch or copy existing documents to quickly use templates or preserve originals.
  • Rich content editing and formatting: Direct your agent to add headers, footers, footnotes, bullet lists, and more—making it easy to update and format documents programmatically.
  • Targeted content manipulation: Have your agent delete specific content ranges, paragraphs, or sections within any document to keep your files up to date.
  • Named range management: Empower your agent to create and manage named ranges for easier referencing, collaboration, and advanced document workflows.
  • Markdown-based document generation: Allow the agent to create new Google Docs directly from markdown content, streamlining content migration from other tools or sources.

Supported Tools & Triggers

Tools
Triggers
Copy Google DocumentTool to create a copy of an existing google document.
Create a documentCreates a new google docs document using the provided title as filename and inserts the initial text at the beginning if non-empty, returning the document's id and metadata (excluding body content).
Create Document MarkdownCreates a new google docs document, optionally initializing it with a title and content provided as markdown text.
Create FooterTool to create a new footer in a google document.
Create FootnoteTool to create a new footnote in a google document.
Create HeaderTool to create a new header in a google document.
Create Named RangeTool to create a new named range in a google document.
Create Paragraph BulletsTool to add bullets to paragraphs within a specified range in a google document.
Delete Content Range in DocumentTool to delete a range of content from a google document.
Delete FooterTool to delete a footer from a google document.
Delete HeaderDeletes the header from the specified section or the default header if no section is specified.
Delete Named RangeTool to delete a named range from a google document.
Delete Paragraph BulletsTool to remove bullets from paragraphs within a specified range in a google document.
Delete TableTool to delete an entire table from a google document.
Delete Table ColumnTool to delete a column from a table in a google document.
Delete Table RowTool to delete a row from a table in a google document.
Get Charts from SpreadsheetTool to retrieve a list of all charts from a specified google sheets spreadsheet.
Get document by idRetrieves an existing google document by its id; will error if the document is not found.
Insert Inline ImageTool to insert an image from a given uri at a specified location in a google document as an inline image.
Insert Page BreakTool to insert a page break into a google document.
Insert Table in Google DocTool to insert a table into a google document.
Insert Table ColumnTool to insert a new column into a table in a google document.
Insert Text into DocumentTool to insert a string of text at a specified location within a google document.
List Charts from SpreadsheetTool to retrieve a list of charts with their ids and metadata from a google sheets spreadsheet.
Replace All Text in DocumentTool to replace all occurrences of a specified text string with another text string throughout a google document.
Replace Image in DocumentTool to replace a specific image in a document with a new image from a uri.
Search DocumentsSearch for google documents using various filters including name, content, date ranges, and more.
Unmerge Table CellsTool to unmerge previously merged cells in a table.
Update Document MarkdownReplaces the entire content of an existing google docs document with new markdown text; requires edit permissions for the document.
Update Document StyleTool to update the overall document style, such as page size, margins, and default text direction.
Update existing documentApplies programmatic edits, such as text insertion, deletion, or formatting, to a specified google doc using the `batchupdate` api method.
Update Table Row StyleTool to update the style of a table row in a google document.

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 Google Docs 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 Google Docs.

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 Google Docs Tool Router session
session = composio.create(
    user_id=user_id,
    toolkits=["googledocs"]
)

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 googledocs.
  • The router checks the user's Google Docs connection and prepares the MCP endpoint.
  • The returned session.mcp.url is the MCP URL that your agent will use to access Google Docs.
  • This approach keeps things lightweight and lets the agent request Google Docs 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 Google Docs. "
        "Help users perform Google Docs 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 Google Docs 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 Google Docs 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 Google Docs.
  • 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 Google Docs 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=["googledocs"]
)
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 Google Docs. "
        "Help users perform Google Docs 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 Google Docs MCP with OpenAI Agents SDK to build a functional AI agent that can interact with Google Docs.

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 Google Docs MCP Agent with another framework

Claude Agent SDK

Claude Agent SDK

Use Google Docs MCP with Claude Agent SDK

Google ADK

Google ADK

Use Google Docs MCP with Google ADK

LangChain

LangChain

Use Google Docs MCP with LangChain

Vercel AI SDK

Vercel AI SDK

Use Google Docs MCP with Vercel AI SDK

Mastra AI

Mastra AI

Use Google Docs MCP with Mastra AI

LlamaIndex

LlamaIndex

Use Google Docs MCP with LlamaIndex

CrewAI

CrewAI

Use Google Docs MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Google Docs MCP?

With a standalone Google Docs MCP server, the agents and LLMs can only access a fixed set of Google Docs tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Google Docs 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 Google Docs tools.

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

Yes, absolutely. You can configure which Google Docs 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 Google Docs 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