returns-refund-policy
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a policy engine that makes automated decisions based on untrusted external data. \n
- Ingestion points: The logic in SKILL.md and the tasks in the evals/ directory ingest data from productTags, customerTags, and orderTags to determine return eligibility and fees. \n
- Boundary markers: No specific delimiters or instructions to ignore embedded commands within the tags are defined in the instructions or logic. \n
- Capability inventory: The skill guides the creation of scripts that can write files, execute shell commands for testing (npm test, npx jest), and automate refund approvals. \n
- Sanitization: The instructions lack requirements for validating or escaping the values in the metadata tags before processing. \n- [EXTERNAL_DOWNLOADS]: The instructions recommend the installation of return management tools from official sources including the Shopify App Store (Loop Returns), BigCommerce App Marketplace (AfterShip), and WordPress.org (ReturnGo).
Audit Metadata