Google Ads MCP
MCP server for Google Ads: analyze campaigns, metrics and keywords with GAQL queries right in the chat with an agent
- 721 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Reads advertising and spend data from a Google Ads account
- Handles a developer token and OAuth credentials
- Runs arbitrary GAQL queries across your accounts
Install
- Works in
Manual install
git clone https://github.com/cohnen/mcp-google-ads.git
cd mcp-google-ads
uv venv .venv
uv pip install -r requirements.txtThere is no PyPI package; the server runs from the cloned repository via the .venv python.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reads advertising and spend data from a Google Ads account
- Handles a developer token and OAuth credentials
- Runs arbitrary GAQL queries across your accounts
This is third-party code. Review the repository files before installing.
What it does
The server connects Google Ads reporting to an agent. It lists accounts, returns campaign and ad metrics for a chosen period and runs arbitrary GAQL queries. Output can come back as a table, JSON or CSV, and the agent reads trends, spots weak points and suggests improvements. Data is read only; the server does not change bids or budgets.
Who it is for. For paid search specialists, marketers and owners who run Google Ads campaigns.
Good fit when
- You want campaign and ad performance analysis in natural language
- You need a quick GAQL query without the Google Ads UI
- You want to find weak keywords and growth points
Not a fit when
- You need to change bids, budgets or launch campaigns; the server is read only
- You have no Google Ads API developer token or account access
Example request
Show the top 10 campaigns over 30 days, flag those with ROAS below 2 and suggest fixesMCP
- Transport
- stdio
- Authentication
- OAuth
Environment variables
GOOGLE_ADS_CREDENTIALS_PATHrequired, secret- Path to the OAuth or service account credentials file
GOOGLE_ADS_DEVELOPER_TOKENrequired, secret- Google Ads API developer token
GOOGLE_ADS_AUTH_TYPE- Auth type: oauth or service_account
GOOGLE_ADS_LOGIN_CUSTOMER_ID- Manager account id for access to several client accounts
Limitations
You need a Google Cloud project with the Google Ads API enabled, a developer token and OAuth or service account credentials, plus a login customer id for a manager account. The developer token is granted after an application, initially in test mode with limits. It installs from the cloned repository through a Python virtual environment; there is no PyPI package. The server only reads data and does not change ad settings.
How to disable. Remove the googleAdsServer block from your MCP client config and restart it, then optionally delete the repository folder with its virtual environment.
FAQ
Where do I get a developer token?
In your Google Ads account under API Center by application. A test token with limits comes first, then a production token after review.
Can it change campaigns?
No. The tools cover reading accounts, metrics and GAQL queries; changing bids and budgets is not included.
Discussion
No questions or reviews yet.
Related
MCP serversGoogle, Meta Ads and GA4 MCP
Google + Meta Ads + GA4 MCP
Remote MCP server for Google Ads, Meta Ads and Google Analytics 4 in one connection, OAuth login without API keys
The experimental official Google Analytics 4 MCP server: reports, funnels and realtime data on request
One MCP server for ad accounts, analytics and CRM: Google Ads, Meta Ads, GA4, TikTok Ads and LinkedIn Ads
An MCP server for Google Ads, Reddit Ads and GA4: the agent reads reports and edits campaigns, and only you turn on real spend
GA4 MCP server: an agent reads Google Analytics 4 reports, schema and real-time data and analyses traffic by playbooks.
Search Console MCP: Google, Bing and GA4 data for agents
Search Console MCP
An MCP server that gives an agent Google Search Console, Bing Webmaster, GA4 and AdSense data with ready SEO checks
README in short
The README walks through setup even without coding experience: Google Ads API access, choosing between OAuth and a service account, and getting a developer token. It covers installing Python and dependencies, using a .env file and configs for Claude Desktop and Cursor. It lists the tools list_accounts, execute_gaql_query, get_campaign_performance, get_ad_performance and run_gaql with GAQL examples. There is a troubleshooting section and test scripts to verify the connection. MIT licensed, maintained by the ixigo team.