Google Ads API v20 MCP Server
Provides comprehensive access to Google Ads API v20, enabling AI assistants to manage campaigns, accounts, assets, and reporting through natural language. It features automatic retry logic, GAQL query support, and advanced functionality for Performance Max and Demand Gen campaigns.
README
Google Ads API v20 MCP Server
A comprehensive Model Context Protocol (MCP) server that provides full access to Google Ads API v20 functionality. This server enables AI assistants to perform any Google Ads operation through natural language commands.
Features
Complete API Coverage
- Account Management: List accounts, get account info, view hierarchy
- Campaign Management: Create, update, pause/resume campaigns with all v20 features
- Ad Group Management: Full CRUD operations for ad groups
- Ad Creation: Responsive search ads, expanded text ads, and more
- Asset Management: Upload and manage images, text assets
- Budget Management: Create and manage shared budgets
- Keyword Management: Add keywords, negative keywords (including Performance Max campaign-level negatives)
- Reporting & Analytics: Custom GAQL queries, performance reports, search terms
- Advanced Features: Recommendations, change history, experiments
Intelligent Features
- Automatic Retry Logic: Handles transient errors with exponential backoff
- Error Documentation: Links to official Google Ads API error documentation
- Partial Failure Handling: Continues processing when some operations fail
- Token Auto-Refresh: Automatically refreshes OAuth tokens
- Self-Rewriting: Can look up documentation and retry failed operations
Installation
# Install with pip
pip install -e .
# Or install dependencies directly
pip install mcp google-ads pydantic httpx tenacity python-dotenv beautifulsoup4 structlog
Configuration
Environment Variables
Create a .env file or set these environment variables:
# OAuth2 Authentication
GOOGLE_ADS_CLIENT_ID=your_client_id
GOOGLE_ADS_CLIENT_SECRET=your_client_secret
GOOGLE_ADS_REFRESH_TOKEN=your_refresh_token
# OR Service Account Authentication
GOOGLE_ADS_SERVICE_ACCOUNT_PATH=/path/to/service-account.json
GOOGLE_ADS_IMPERSONATED_EMAIL=user@example.com # Optional
# Required for all auth methods
GOOGLE_ADS_DEVELOPER_TOKEN=your_developer_token
GOOGLE_ADS_LOGIN_CUSTOMER_ID=1234567890 # Manager account ID if applicable
Configuration File
Alternatively, create a config file at ~/.config/google-ads-mcp/config.json:
{
"client_id": "your_client_id",
"client_secret": "your_client_secret",
"refresh_token": "your_refresh_token",
"developer_token": "your_developer_token",
"login_customer_id": "1234567890"
}
MCP Configuration
Add to your Claude Desktop config (~/.config/claude/mcp.json):
{
"mcpServers": {
"google-ads": {
"command": "python",
"args": ["-m", "google-ads-mcp"],
"env": {
"GOOGLE_ADS_DEVELOPER_TOKEN": "your_token",
"GOOGLE_ADS_CLIENT_ID": "your_client_id",
"GOOGLE_ADS_CLIENT_SECRET": "your_secret",
"GOOGLE_ADS_REFRESH_TOKEN": "your_refresh_token"
}
}
}
}
Usage Examples
Basic Operations
# List all accounts
Use the list_accounts tool
# Create a campaign
Use create_campaign with customer_id="1234567890", name="Summer Sale 2025",
budget_amount=100.0, campaign_type="SEARCH"
# Get campaign performance
Use get_campaign_performance with customer_id="1234567890", date_range="LAST_30_DAYS"
Advanced Queries
# Run custom GAQL query
Use run_gaql_query with query:
SELECT campaign.name, metrics.clicks, metrics.conversions
FROM campaign
WHERE metrics.impressions > 1000
AND segments.date DURING LAST_7_DAYS
ORDER BY metrics.clicks DESC
Performance Max Features (v20)
# Add negative keywords to Performance Max campaign
Use add_negative_keywords with customer_id="1234567890",
campaign_id="123", keywords=["cheap", "discount", "free"]
Error Handling
The server provides detailed error information:
- Error type and code
- Human-readable message
- Whether the error is retryable
- Link to official documentation
- Suggestions for fixing common errors
Development
Running Tests
pytest tests/
Adding New Tools
- Add tool definition to
_register_tools()intools.py - Implement the handler method
- Update documentation
Debugging
# Run with debug logging
export LOG_LEVEL=DEBUG
python -m google-ads-mcp
Security Notes
- Never commit credentials to version control
- Use service accounts for production environments
- Enable 2FA on Google Ads accounts
- Regularly rotate refresh tokens
- Monitor API usage and set alerts
API Version Support
This server is built for Google Ads API v20 (released June 2025) and includes:
- Campaign-level negative keywords for Performance Max
- Enhanced Demand Gen reporting with channel segmentation
- Platform comparable conversions
- All v20-specific features and improvements
License
MIT License - See LICENSE file for details
Contributing
Contributions are welcome! Please:
- Fork the repository
- Create a feature branch
- Add tests for new functionality
- Submit a pull request
Support
For issues and questions:
- Check the Google Ads API documentation
- Review error messages and documentation links
- Open an issue on GitHub
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。