sirv-ecommerce
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes data from external sources such as Shopify product catalogs and remote image URLs, creating a surface for indirect prompt injection. * Ingestion points: Shopify product information is imported via
sirv_sync_shopify_products, and external image URLs are processed throughsirv_execute_workflowandsirv_run_supplier_autofix. * Boundary markers: The instructions do not specify the use of delimiters or ignore-embedded-instruction warnings when handling external metadata. * Capability inventory: The agent has the capability to delete Shopify product images and update catalog assets. * Sanitization: No specific sanitization or filtering logic is provided for the imported product descriptions or image metadata. - [COMMAND_EXECUTION]: The skill includes tool parameters capable of performing destructive operations on external e-commerce assets. * Evidence: The
sirv_push_image_to_shopifytool includes areplace_allstrategy that deletes every image associated with a Shopify product before adding new ones. * Context: The documentation explicitly labels this strategy as destructive and directs the agent to obtain user confirmation before execution, effectively mitigating the risk of accidental data loss.
Audit Metadata