giuseppe-coco/Google-Workspace-MCP-Server

A Model Context Protocol (MCP) server that acts as a secure bridge between your personal Google Workspace account (Gmail, Calendar, etc.) and any MCP-compatible AI client, such as Claude Desktop.
Features
* **Google Calendar**:
* Effortlessly list and search for events on your primary calendar within a specific date range.
* Create new events with detailed information like title, description, start, and end times.
* Update existing events, allowing for partial modifications such as changing the title or time.
* Delete events directly from your calendar.
* **Gmail**:
* Read the content of your most recent email to stay up-to-date.
* Search for specific emails by their subject line to find important conversations.
* Compose and send new emails directly from your account.
* **Google Drive**:
* Search for files and folders using powerful query strings.
* Create new Google Docs with a specified title and initial content.
* Update the entire content of an existing Google Doc.
* Manage your files by moving them to the bin or deleting them permanently.
Getting Started
Follow these steps to set up the server and run the example AI agent.
Prerequisites
* Python 3.9+ and uv (or pip).
* A Google Cloud project with the necessary APIs enabled.
* Claude Desktop.
Step 1: Configure your Google Cloud Project
You need to authorize this application to access your Google data. This is a one-time setup.
1. **Go to the Google Cloud Console**: https://console.cloud.google.com/
2. **Create a new project** (or use an existing one).
3. **Enable APIs**:
* Go to "APIs & Services" -> "Library".
* Search for and **Enable** the **Gmail API**.
* Search for and **Enable** the **Google Calendar API**.
* Search for and **Enable** the **Google Drive API**.
4. **Create OAuth Credentials**:
* Go to "APIs & Services" -> "Credentials".
* Click "Create Credentials" -> "OAuth client ID".
* If prompted, configure the "OAuth consent screen". Choose **External** and provide a name for the app. You can skip most other fields for personal use. Add your Google account email as a Test User.
* For "Application type", select **Desktop app**.
* Give it a name (e.g., "GSuite MCP Client").
5. **Download Credentials**:
* After creating the client ID, click the "Download JSON" icon.
* Rename the downloaded file to client_secrets.json and place it in the **root directory of this project**.
Step 2: Install Dependencies
Clone this repository and install the required Python packages for both the server and the client.
Step 3: Run the One-Time Authorization
Before you can run the server, you need to authorize it with your Google account. Run the get_credentials.py script from your terminal:
* This will open a browser window.
* Log in with your Google account and grant the requested permissions.
* After you approve, the script will automatically create a token.json file in the project directory. This file stores your authorization tokens so you don't have to log in every time.
Step 4: Set Up and Run the AI Agent Client