google-ads-search-terms
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill ingests search terms via the
google-ads-mcptool (specifically thesearch_term_viewtable) and manual user exports. These terms are strings generated by arbitrary users on the internet. - Boundary markers: The instructions lack clear delimiters or "ignore instructions" directives to prevent the agent from accidentally obeying commands embedded within the search term strings.
- Capability inventory: The skill is authorized to write new files and update existing memory files within the
workspace/directory (e.g.,workspace/ads/negatives.md,workspace/ads/findings.md). - Sanitization: There is no evidence of sanitization, escaping, or filtering applied to the search term data before it is processed by the LLM and written to the workspace.
Audit Metadata