google-ads-api-mcp-setup

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill is a legitimate setup guide for the official Google Ads MCP server with no malicious signs, but its Cloud Run example deploys with --allow-unauthenticated and passes OAuth secrets as environment variables, which exposes the user's Google Ads access to anyone with the URL, a careless weakening of protection beyond what the task needs.