google-ads-monitor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from Google Ads API responses (such as campaign names, ad group names, and ad IDs) which represents an attack surface for indirect prompt injection if an attacker controls the ad account contents.\n
- Ingestion points: Google Ads data is retrieved using
mcp__cogny__google_ads__tool_execute_gaqlas described inSKILL.md.\n - Boundary markers: The instructions do not specify explicit delimiters or markers to separate untrusted campaign/ad labels from internal report templates.\n
- Capability inventory: The skill is authorized to write to context nodes (
mcp__cogny__write_context_node) and generate system findings (mcp__cogny__create_finding).\n - Sanitization: No explicit sanitization or validation logic is applied to the text fields retrieved from the Google Ads service before they are interpolated into the user-facing report.
Audit Metadata