google-shopping-search
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves product listings from Google Shopping, an external source that is not under the vendor's direct control. Malicious listings could theoretically contain instructions designed to influence the agent when it summarizes the search results.\n
- Ingestion points: API response body processed in
scripts/search_google_shopping.sh.\n - Boundary markers: The script does not use specific markers to delimit untrusted data for the agent.\n
- Capability inventory: The skill can execute bash scripts, perform network requests via
curl, and write to the local file system.\n - Sanitization: The script uses Python to validate that the response is a valid JSON object before processing it.\n- [COMMAND_EXECUTION]: The skill utilizes a bash script to coordinate its tasks, including fetching data and running inline Python code. It safely manages search parameters by passing them as environment variables to Python.\n- [EXTERNAL_DOWNLOADS]: The skill performs network requests to
ai-factory.frevana.comto retrieve search data, which is its primary function.
Audit Metadata