google-analytics

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow ingests the outsider-authored user’s natural-language query at runtime to determine parameters (date range, dimensions, filters) and then calls the GA4 MCP tools (runReport, getPageViews, getEvents, getUserBehavior) with those derived free-text inputs.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup runs an external MCP server via "npx mcp-server-google-analytics" (a required runtime step) which fetches and executes the package code corresponding to https://github.com/ruchernchong/mcp-server-google-analytics, so this URL represents a runtime external dependency that executes remote code.