Google has open-sourced a Mannequin Context Protocol (MCP) server that exposes read-only entry to the Google Adverts API for agentic and LLM functions. The repository googleads/google-ads-mcp implements an MCP server in Python that surfaces two instruments in the present day: search (GAQL queries over Adverts accounts) and list_accessible_customers (enumeration of buyer assets). It contains setup by way of pipx, Google Adverts developer tokens, OAuth2 scopes (https://www.googleapis.com/auth/adwords), and Gemini CLI / Code Help integration by means of a normal MCP shopper configuration. The undertaking is labeled “Experimental.”
So, why it issues?
MCP is rising as a standard interface for wiring fashions to exterior methods. By delivery a reference server for the Adverts API, Google lowers the combination price for LLM brokers that want marketing campaign telemetry, funds pacing, and efficiency diagnostics with out bespoke SDK glue.
The way it works? (developer view)
- Protocol: MCP standardizes “instruments” that fashions can invoke with typed parameters and responses. The Adverts MCP server advertises instruments mapped to Google Adverts API operations; MCP shoppers (Gemini CLI/Code Help, others) uncover and name them throughout a session.
- Auth & scopes: You allow the Google Adverts API in a Cloud undertaking, acquire a developer token, and configure Utility Default Credentials or the Adverts Python shopper. Required scope is
adwords. For manager-account hierarchies, set a login buyer ID. - Consumer wiring: Add a
~/.gemini/settings.jsonentry pointing to the MCP server invocation (pipx run git+https://github.com/googleads/google-ads-mcp.git google-ads-mcp) and move credentials by way of env vars. Then question by way of/mcpin Gemini or by prompting for campaigns, efficiency, and many others.
Ecosystem sign
Google’s server arrives amid broader MCP adoption throughout distributors and open-source shoppers, reinforcing MCP as a practical path to agent-to-SaaS interoperability. For PPC and development groups experimenting with agentic workflows, the reference server is a low-friction strategy to validate LLM-assisted QA, anomaly triage, and weekly reporting with out granting write privileges.
Key Takeaways
- Google open-sourced a read-only Google Adverts API MCP server, showcasing two instruments:
search(GAQL) andlist_accessible_customers. - Implementation particulars: Python undertaking on GitHub (
googleads/google-ads-mcp), Apache-2.0 license, marked Experimental; set up/run by way ofpipxand configure OAuth2 with thehttps://www.googleapis.com/auth/adwordsscope (dev token + elective login-customer ID). - Works with MCP-compatible shoppers (e.g., Gemini CLI / Code Help) so brokers can difficulty GAQL queries and analyze Adverts accounts by means of natural-language prompts.
Conclusion
In sensible phrases, Google’s open-sourced Google Adverts API MCP server provides groups a standards-based, read-only path for LLM brokers to run GAQL queries in opposition to Adverts accounts with out bespoke SDK wiring. The Apache-licensed repo is marked experimental, exposes search and list_accessible_customers, and integrates with MCP shoppers like Gemini CLI/Code Help; manufacturing use ought to account for OAuth scope (adwords), developer token administration, and the data-exposure caveat famous within the README.
Try the GitHub Web page and technical weblog. Be at liberty to take a look at our GitHub Web page for Tutorials, Codes and Notebooks. Additionally, be happy to observe us on Twitter and don’t overlook to hitch our 100k+ ML SubReddit and Subscribe to our Publication. Wait! are you on telegram? now you may be a part of us on telegram as effectively.
Michal Sutter is an information science skilled with a Grasp of Science in Information Science from the College of Padova. With a strong basis in statistical evaluation, machine studying, and information engineering, Michal excels at remodeling complicated datasets into actionable insights.
