Amazon Seller
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from Amazon, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill ingests competitor pricing, product listing data, and customer details (e.g.,
buyer_name) from the Amazon Marketplace and Advertising APIs. - Boundary markers: Templates for product listings and customer messages do not include explicit instructions to use delimiters or "ignore embedded instructions" warnings when interpolating external data.
- Capability inventory: The skill possesses significant capabilities, including updating live product listings, modifying advertising campaigns, and sending automated messages to customers.
- Sanitization: The skill does not define specific validation or sanitization procedures for data retrieved from external sources before it is used in automated actions.
Audit Metadata