amz-return-reduction
Pass
Audited by Gen Agent Trust Hub on Jun 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely text-based and instructional. It specifically states that it 'Works with zero tools', meaning it does not require or use any external scripts, command execution, or environment permissions.
- [SAFE]: While the skill is designed to process untrusted data (customer reviews and return comments), it poses no risk of indirect prompt injection because it lacks any capabilities that could be exploited (e.g., it cannot write to files, execute commands, or make network requests).
- [SAFE]: The inclusion of a link to a WhatsApp community in the footer is for author attribution and community building. It is not an automated network call and does not facilitate data exfiltration from the agent's environment.
- [SAFE]: The logic provided for calculating fees (Returns Processing Fee and Refund Administration Fee) is based on standard Amazon Seller Central policies and involves manual data entry by the user rather than automated credential access.
Audit Metadata