google-ads

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

• Third-party content exposure detected (high risk: 0.90). The skill's mandatory Phase 2 research instructs the agent to "Scan the site end-to-end" and "Inspect SERPs, competitors, and themes," which requires fetching and interpreting content from public websites and competitor pages (untrusted third‑party sources) to drive campaign decisions.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill is explicitly for creating and managing Google Ads campaigns via dedicated integration tools (e.g., google_ads_preview_blueprint, google_ads_create_from_blueprint, google_ads_create_from_pmax_blueprint). The blueprints include a required budget field ("budget_amount_micros"), rules for budgets, and workflows to create and later activate campaigns (which control ad spend). It therefore provides specific API-level capabilities to set and deploy ad budgets/bidding strategies (directly affecting spend), not just generic browsing or reporting. This meets the "managing ad spend budgets (API to update budget)" criterion for direct financial execution.