google-ads-api-mcp-setup

The google-ads-api-mcp-setup skill is a structured installation and configuration guide for the official open-source Google Ads Model Context Protocol (MCP) Server. It solves the setup friction of connecting an AI assistant (such as Gemini, Claude Code, or Cursor) to a Google Ads account so users can query campaigns and retrieve reporting metrics in natural language. Rather than shipping code, it directs the agent through a repeatable, safety-checked installation flow that points only at the official package and repository.

The skill lays out prerequisites (Python 3.12+ and pipx, plus outbound HTTPS to the Google Ads API and PyPI) and a credential checklist covering the five standard Google Ads API parameters: Developer Token, OAuth2 Client ID and Client Secret, OAuth2 Refresh Token, and Customer IDs, with formatting rules (digits only, no hyphens) and a login Customer ID for MCC hierarchies. If credentials are missing it defers to a companion quickstart skill. It then walks through verifying and installing prerequisites with OS-specific commands (Homebrew, scoop, or apt, plus pipx ensurepath), installing the stable google-ads-mcp package from PyPI by default (GitHub development version only on explicit request), verifying with a --help command, troubleshooting PATH issues via the pipx bin directory, and explaining the stdio transport for the MCP connection.

It targets developers and marketing engineers who want to wire an AI assistant to Google Ads reporting. The credential handling follows standard official-tooling practice: it documents the required tokens and OAuth values but does not exfiltrate them, and it explicitly steers users to the vetted official server rather than third-party ones.