marketplace-advertising
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow for harvesting search terms from external marketplace reports to automate keyword targeting. This represents a potential surface for indirect prompt injection if the agent interprets search queries as instructions.
- Ingestion points: Search Term Reports retrieved via the Amazon Advertising API (SKILL.md, Step 2).
- Boundary markers: Absent; the instructions do not mention using delimiters or explicit 'ignore instructions' warnings when processing search term strings.
- Capability inventory: The skill describes the use of API calls to create new keywords and negative keywords in active ad campaigns (SKILL.md, Step 2 and Step 6).
- Sanitization: Absent; search terms are filtered by performance metrics but the strings themselves are not sanitized or validated for instructional content before being used in API calls.
- Self-referential content: The skill metadata explicitly claims 'risk: safe', which is a self-referential safety claim that must be evaluated independently during security analysis (SKILL.md frontmatter).
Audit Metadata