ecommerce-ozon-category-products
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted product metadata from the Ozon marketplace, which could contain malicious instructions intended to influence agent behavior.\n
- Ingestion points: The
scripts/ozon_category_products.pyscript fetches data from the external API and saves the results to JSON files in the project directory.\n - Boundary markers: The skill documentation does not provide delimiters or instructions for the agent to treat the product data as non-executable text, leaving it vulnerable to embedded instructions.\n
- Capability inventory: The skill enables the agent to execute a Python script and interact with the local filesystem to read the saved marketplace data.\n
- Sanitization: There is no evidence of sanitization or escaping of the external product titles, brand names, or other text fields before they are provided to the agent.
Audit Metadata