google-ads-rsas
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests search term data from external Google Ads reports. This represents a surface where attacker-controlled data (search queries) could contain instructions intended to influence the agent. However, the risk is mitigated by the skill's narrow scope of generating ad drafts and its lack of high-privilege execution capabilities.
- Ingestion points: Search term data is pulled via the
searchtool ongoogle-ads-mcp(SKILL.md). - Boundary markers: None explicitly defined in instructions.
- Capability inventory: File write operations to the local
workspace/directory (SKILL.md). - Sanitization: No explicit sanitization or filtering of external content is defined.
Audit Metadata