amazon-listing-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a bundled bash script,
scripts/fetch-listing.sh, to automate the retrieval of product metadata from Amazon. This script is invoked to support the audit and optimization workflows. - [EXTERNAL_DOWNLOADS]: The skill performs network requests using
curlto fetch product data from various official Amazon marketplace domains (e.g.,amazon.com,amazon.co.uk,amazon.de). These operations are consistent with the skill's stated purpose of Amazon listing optimization. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze data from external Amazon listings, which could contain adversarial instructions hidden in product titles or descriptions.
- Ingestion points: Data is fetched from Amazon product pages via the
fetch-listing.shscript, which extracts titles, bullet points, and descriptions. - Boundary markers: The skill instructions do not specify explicit delimiters or "ignore instructions" tags when processing the fetched listing content.
- Capability inventory: The skill environment allows for shell script execution (
fetch-listing.sh) and network operations viacurl. - Sanitization: The
fetch-listing.shscript usesgrepandsedto isolate specific listing elements, which provides a layer of structural filtering before the data is presented to the agent.
Audit Metadata