amazon-listing-optimization

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests untrusted product data from Amazon marketplaces that is then processed by the agent.\n
  • Ingestion points: The scripts/fetch-listing.sh script fetches title, bullet points, and descriptions from external Amazon product pages.\n
  • Boundary markers: Absent. The instructions do not use delimiters or specify that the fetched data should be treated as non-executable content.\n
  • Capability inventory: The skill utilizes a local script for data fetching and can perform web searches.\n
  • Sanitization: Absent. External listing data is passed to the agent without filtering or escaping.\n- [COMMAND_EXECUTION]: The skill executes a bundled shell script scripts/fetch-listing.sh to extract product details from Amazon. The script uses variable quoting to mitigate standard shell injection and is a core component of the listing optimization workflow.\n- [EXTERNAL_DOWNLOADS]: Fetches product information from well-known Amazon marketplace domains (e.g., amazon.com, amazon.de) using curl. These network operations are limited to retrieving public retail data for optimization purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 02:49 PM