agency-search-query-analyst
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and analyzing untrusted external data (raw search queries from users).
- Ingestion points: The skill explicitly instructs the agent to "Pull live search term reports" and "Run n-gram analysis at scale on actual query data" from external Google Ads API integrations or MCP tools.
- Boundary markers: There are no instructions provided to delimit search query data from agent instructions, nor are there warnings to the agent to disregard instructions potentially embedded within those queries.
- Capability inventory: The skill includes high-impact capabilities such as "Push negative keyword changes back to the account," which could be manipulated if an attacker can influence the agent's interpretation of query data.
- Sanitization: The skill lacks any requirement or guidance for sanitizing, escaping, or validating the external search queries before they are processed for intent classification or keyword generation.
Audit Metadata