shop-app
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external merchants (product details) and third-party stores (policies). It includes explicit instructions for the agent to ignore any commands or instructions embedded within this merchant-supplied data, which serves as a mitigation against indirect prompt injection.
- [COMMAND_EXECUTION]: The skill utilizes the
curlcommand to interact with the Shop.app API and thebase64utility for image processing. These operations are restricted to the service's official domain and are used for intended functionality.
Audit Metadata