A FastMCP-powered Model Context Protocol server for Google Ads API integration with automatic OAuth 2.0 authentication

Connect Google Ads API directly to Claude Desktop and other MCP clients with seamless OAuth 2.0 authentication, automatic token refresh, GAQL querying, and keyword research capabilities.

For a simpler setup experience, we offer ready-to-use installers:

👉 Download installer - https://gomarble.ai/mcp

👉 Slack Community - AI in Ads

👉 Facebook Ads MCP - Facebook Ads MCP

• 🔐 Automatic OAuth 2.0 - One-time browser authentication with auto-refresh
• 🔄 Smart Token Management - Handles expired tokens automatically
• 📊 GAQL Query Execution - Run any Google Ads Query Language queries
• 🏢 Account Management - List and manage Google Ads accounts
• 🔍 Keyword Research - Generate keyword ideas with search volume data
• 🚀 FastMCP Framework - Built on the modern MCP standard
• 🖥️ Claude Desktop Ready - Direct integration with Claude Desktop
• 🛡️ Secure Local Storage - Tokens stored locally, never exposed

Note: All tools automatically handle authentication - no token parameters required!

Before setting up the MCP server, you'll need:

• Python 3.10+ installed
• A Google Cloud Platform account
• A Google Ads account with API access

1. Go to Google Cloud Console
2. Create a new project:
   • Click "Select a project" → "New Project"
   • Enter project name (e.g., "Google Ads MCP")
   • Click "Create"

1. In your Google Cloud Console:
   • Go to "APIs & Services" → "Library"
   • Search for "Google Ads API"
   • Click on it and press "Enable"

1. Go to "APIs & Services" → "Credentials"
2. Click "+ CREATE CREDENTIALS" → "OAuth 2.0 Client ID"
3. Configure consent screen (if first time):
   • Click "Configure Consent Screen"
   • Choose "External" (unless you have Google Workspace)
   • Fill required fields:
     • App name: "Google Ads MCP"
     • User support email: Your email
     • Developer contact: Your email
   • Click "Save and Continue" through all steps
4. Create OAuth Client:
   • Application type: "Desktop application"
   • Name: "Google Ads MCP Client"
   • Click "Create"
5. Download credentials:
   • Click "Download JSON" button
   • Save file as client_secret_[long-string].json in your project directory

1. Sign in to Google Ads
2. Go to Tools & Settings (wrench icon in top navigation)
3. Under "Setup", click "API Center"
4. Accept Terms of Service if prompted
5. Click "Apply for token"
6. Fill out application form:
   • Describe your use case (e.g., "MCP integration for campaign analysis")
   • Provide technical details about your implementation
7. Submit and wait for approval (usually 1-3 business days)

Note: You'll initially get a test token with limited functionality. After testing, you can apply for production access.

Once approved:

1. Return to API Center in Google Ads
2. Copy your Developer Token (format: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX)

# Clone the repository
git clone https://github.com/yourusername/google-ads-mcp-server.git
cd google-ads-mcp-server

# Create virtual environment (recommended)
python3 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Create a .env file in your project directory:

# Copy the example file
cp .env.example .env

Edit .env with your credentials:

# Required: Google Ads API Developer Token
GOOGLE_ADS_DEVELOPER_TOKEN=your_developer_token_here

# Required: Path to OAuth credentials JSON file (downloaded from Google Cloud)
GOOGLE_ADS_OAUTH_CONFIG_PATH=/full/path/to/your/client_secret_file.json

Example .env file:

GOOGLE_ADS_DEVELOPER_TOKEN=ABCDEFG1234567890
GOOGLE_ADS_OAUTH_CONFIG_PATH=/Users/john/google-ads-mcp/client_secret_138737274875-abc123.apps.googleusercontent.com.json

Find your Claude Desktop configuration file:

macOS:

~/Library/Application Support/Claude/claude_desktop_config.json

Windows:

%APPDATA%\Claude\claude_desktop_config.json

Edit the configuration file and add your Google Ads MCP server:

{
  "mcpServers": {
    "google-ads": {
      "command": "/full/path/to/your/project/.venv/bin/python",
      "args": [
        "/full/path/to/your/project/server.py"
      ]
    }
  }
}

Real Example:

{
  "mcpServers": {
    "google-ads": {
      "command": "/Users/marble-dev-01/workspace/google_ads_with_fastmcp/.venv/bin/python",
      "args": [
        "/Users/marble-dev-01/workspace/google_ads_with_fastmcp/server.py"
      ]
    }
  }
}

Important:

• Use absolute paths for all file locations
• On Windows, use forward slashes / or double backslashes \\ in paths
• Replace your_developer_token_here with your actual developer token

Close and restart Claude Desktop to load the new configuration.

1. Open Claude Desktop
2. Try any Google Ads command, for example:

   "List all my Google Ads accounts"
   

1. Browser opens automatically to Google OAuth page
2. Sign in with your Google account (the one with Google Ads access)
3. Grant permissions by clicking "Allow"
4. Browser shows success page
5. Return to Claude - your command will complete automatically!

After authentication, you should see:

• A google_ads_token.json file created in your project directory
• Your Google Ads accounts listed in Claude's response

"List all my Google Ads accounts"

"Show me the account details and which ones have active campaigns"

"Show me campaign performance for account 1234567890 in the last 30 days"

"Get conversion data for all campaigns in the last week"

"Which campaigns have the highest cost per conversion?"

"Generate keyword ideas for 'digital marketing' using account 1234567890"

"Find keyword opportunities for 'AI automation' with search volume data"

"Research keywords for the page https://example.com/services"

"Run this GAQL query for account 1234567890:
SELECT campaign.name, metrics.clicks, metrics.cost_micros 
FROM campaign 
WHERE segments.date DURING LAST_7_DAYS"

"Get keyword performance data:
SELECT ad_group_criterion.keyword.text, metrics.ctr, metrics.average_cpc
FROM keyword_view 
WHERE metrics.impressions > 100"

SELECT 
  campaign.id,
  campaign.name, 
  metrics.clicks, 
  metrics.impressions,
  metrics.cost_micros,
  metrics.conversions,
  metrics.conversions_value
FROM campaign 
WHERE segments.date DURING LAST_30_DAYS
ORDER BY metrics.cost_micros DESC

SELECT 
  campaign.name,
  ad_group_criterion.keyword.text, 
  ad_group_criterion.keyword.match_type,
  metrics.ctr,
  metrics.average_cpc,
  metrics.quality_score
FROM keyword_view 
WHERE segments.date DURING LAST_7_DAYS
  AND metrics.impressions > 100
ORDER BY metrics.conversions DESC

SELECT 
  campaign.name,
  segments.device,
  metrics.clicks,
  metrics.cost_micros,
  metrics.conversions
FROM campaign
WHERE segments.date DURING LAST_30_DAYS
  AND campaign.status = 'ENABLED'

google-ads-mcp-server/
├── server.py                           # Main MCP server
├── oauth/
│   ├── __init__.py                     # Package initialization
│   └── google_auth.py                  # OAuth authentication logic
├── google_ads_token.json               # Auto-generated token storage (gitignored)
├── client_secret_[long-string].json    # Your OAuth credentials (gitignored)
├── .env                                # Environment variables (gitignored)
├── .env.example                        # Environment template
├── .gitignore                          # Git ignore file
├── requirements.txt                    # Python dependencies
├── LICENSE                             # MIT License
└── README.md                           # This file

• ✅ Credential files are gitignored - Never committed to version control
• ✅ Local token storage - Tokens stored in google_ads_token.json locally
• ✅ Environment variables - Sensitive data in .env file
• ✅ Automatic refresh - Minimal token exposure time

# Set secure permissions for sensitive files
chmod 600 .env
chmod 600 google_ads_token.json
chmod 600 client_secret_*.json

1. Use environment variables instead of .env files in production
2. Implement rate limiting to respect API quotas
3. Monitor API usage in Google Cloud Console
4. Secure token storage with proper access controls
5. Regular token rotation for enhanced security

Enable detailed logging for troubleshooting:

# Add to server.py for debugging
import logging
logging.basicConfig(level=logging.DEBUG)

If you encounter issues:

1. Check the error message carefully - it usually indicates the exact problem
2. Verify all file paths are absolute and correct
3. Ensure environment variables are properly set
4. Check Google Cloud Console for API quotas and billing
5. Restart Claude Desktop after any configuration changes

For web deployment or remote access:

# Start server in HTTP mode
python3 server.py --http

Claude Desktop config for HTTP:

{
  "mcpServers": {
    "google-ads": {
      "url": "http://127.0.0.1:8000/mcp"
    }
  }
}

Modify token storage location in oauth/google_auth.py:

# Custom token file location
def get_token_path():
    return "/custom/secure/path/google_ads_token.json"

For managing multiple accounts under an MCC:

# Add to .env file
GOOGLE_ADS_LOGIN_CUSTOMER_ID=123-456-7890

We welcome contributions! Here's how to get started:

# Fork and clone the repository
git clone https://github.com/yourusername/google-ads-mcp-server.git
cd google-ads-mcp-server

# Create development environment
python3 -m venv .venv
source .venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Set up development environment
cp .env.example .env
# Add your development credentials to .env

1. Create a feature branch: git checkout -b feature/amazing-feature
2. Make your changes with appropriate tests
3. Test thoroughly with different account configurations
4. Update documentation as needed
5. Commit changes: git commit -m 'Add amazing feature'
6. Push to branch: git push origin feature/amazing-feature
7. Open a Pull Request with detailed description

# Test authentication flow
python3 server.py --test-auth

# Test API connectivity
python3 -c "
from oauth.google_auth import get_oauth_credentials
creds = get_oauth_credentials()
print('✅ Authentication successful!')
"

# Test with Claude Desktop
# Add your server to Claude config and test various commands

• Basic access: 15,000 operations per day
• Standard access: 40,000 operations per day
• Request rate: 1,600 requests per minute per developer token

1. Cache results when possible to reduce API calls
2. Use date ranges to limit data volume
3. Batch requests when supported
4. Monitor usage in Google Cloud Console
5. Implement retry logic for rate limit errors

# Monitor usage in Google Cloud Console
# Go to APIs & Services → Quotas
# Search for "Google Ads API" to see current usage

This project is licensed under the MIT License - see the LICENSE file for details.

Copyright (c) 2025 Google Ads MCP Server Contributors

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

• 🔄 Enhanced keyword research with competitor analysis
• 📊 Built-in data visualization with charts and graphs
• 🤖 AI-powered optimization suggestions
• 📝 Campaign creation and management tools
• 🔍 Advanced reporting capabilities
• 🌐 Multi-language support

Made with ❤️ for the MCP community

Connect your Google Ads data directly to AI assistants and unlock powerful advertising insights through natural language conversations.