ecommerce-1688-product-billboard

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/1688_product_billboard.py performs network requests using the standard urllib library and writes cache files to the local file system.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves product data from the NexScope API proxy, which is a core function of the skill and targets a vendor-controlled endpoint.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a data ingestion surface by processing third-party product titles and seller information from the 1688 marketplace. 1. Ingestion points: Product titles, shop names, and category hierarchies in the API response. 2. Boundary markers: No explicit isolation markers are used in the prompt instructions to delimit external product content. 3. Capability inventory: The script can perform network operations and write files to the local disk. 4. Sanitization: No sanitization or filtering of the external marketplace text is implemented prior to presentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:12 AM
Security Audit — agent-trust-hub — ecommerce-1688-product-billboard