ecommerce-shopee-product-search
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Python script (
scripts/shopee_product_search.py) to handle API communication, implement local caching, and save search results to the filesystem. The script is invoked with JSON parameters derived from user input. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves product metadata (titles, descriptions, category structures) from the Shopee platform via a proxy API, creating an attack surface where malicious content embedded in product listings could influence the agent's behavior.
- Ingestion points: The agent ingests data through the standard output of the
shopee_product_search.pyscript, which returns product information from an external data source. - Boundary markers: There are no specific instructions or delimiters defined to help the agent distinguish between its instructions and the potentially untrusted product data.
- Capability inventory: The agent has the capability to execute the search script and read/write files in the
nexscopedirectory within its working environment. - Sanitization: The Python script does not perform sanitization or filtering of the Shopee product content before passing it to the agent's context.
Audit Metadata