amz-brand-tailored-promotions
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided business data such as brand goals, audience sizes, and product margins.
- Ingestion points: Untrusted data enters the agent context through the "Collect inputs" phase (Step 1).
- Boundary markers: The instructions do not implement specific delimiters or warnings to ignore instructions embedded within the user's business data.
- Capability inventory: The skill is restricted to text processing and has no access to tools, file system operations, or network requests.
- Sanitization: There is no evidence of input validation or sanitization routines.
- [SAFE]: The skill includes an external link to a community WhatsApp group for informational purposes.
- Evidence: "Join here: https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY"
Audit Metadata