ecommerce-ozon-seller-products
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs legitimate network operations to fetch e-commerce data from the vendor's API proxy (
NEXSCOPE_PROXY_BASE). - [SAFE]: Sensitive configuration (API keys) is handled via the
NEXSCOPE_API_KEYenvironment variable, following security best practices. - [SAFE]: The Python script
scripts/ozon_seller_products.pyperforms localized file writes to store session-specific data and cache results, with appropriate error handling and fallback to temporary directories. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from a marketplace API. While this presents an ingestion surface for untrusted content (e.g., product titles), the script handles data as structured JSON and the agent is instructed to present the data in tables, minimizing the risk of instruction execution.
- Ingestion points:
scripts/ozon_seller_products.py(viaurlopencalling the Ozon Seller Products API) - Boundary markers: Absent
- Capability inventory: File-write (
os.makedirs,open(..., "w")) and Network-op (urlopen) - Sanitization: Absent (standard JSON parsing only)
Audit Metadata