demand-gen-audit
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the Google Ads platform (such as campaign names, ad group names, and ad creative metadata) which could potentially contain malicious instructions aimed at influencing the agent's behavior.
- Ingestion points: Data enters the agent context through the
mcp__cogny__google_ads__tool_execute_gaqltool as seen in the data retrieval steps inSKILL.md. - Boundary markers: No specific delimiters or "ignore instructions" warnings are defined in the instructions to isolate the data returned by the Google Ads API tools.
- Capability inventory: The skill utilizes tools like
WebFetch,WebSearch, andmcp__cogny__create_findingwhich provide a functional surface for an injection to trigger external requests or data recording. - Sanitization: The skill's instructions do not specify sanitization or validation of the content retrieved from the external advertising account.
Audit Metadata