ecommerce-temu-store-source-query

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes store data from an external platform (Temu), creating a surface for indirect prompt injection if that data contains malicious instructions. \n
  • Ingestion points: The script scripts/temu_store_source_query.py retrieves store details including names and categories from a remote gateway. \n
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential commands embedded within the fetched store data. \n
  • Capability inventory: The skill possesses capabilities for network communication (urllib.request) and local file system writes (open().write() to the nexscope/ directory). \n
  • Sanitization: The script saves the raw JSON response and prints contents without performing sanitization on text fields that could contain adversarial content. \n
  • [DATA_EXFILTRATION]: The skill script performs network requests to an external API gateway. \n
  • Evidence: scripts/temu_store_source_query.py uses urllib.request.urlopen to send POST requests to a URL defined by the NEXSCOPE_PROXY_BASE environment variable. \n
  • Context: This network activity is documented as functional for the purpose of retrieving third-party data from the vendor's infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:13 AM
Security Audit — agent-trust-hub — ecommerce-temu-store-source-query