buy-box-loss-root-cause
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for the agent to synthesize reports based on structured Amazon business data. It does not perform any dangerous operations such as arbitrary command execution, persistence, or privilege escalation.- [DATA_EXPOSURE & EXFILTRATION]: The skill accesses Amazon profit and inventory data via the vendor's infrastructure (mcp.datadoe.com). This data access is aligned with the skill's primary purpose and is restricted to the platform's standard export tools. No exfiltration to unauthorized third-party services was observed.- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it processes external table data (e.g., amazon_profit_by_sku_and_date). Evidence chain:
- Ingestion points: The skill pulls data from 'amazon_profit_by_sku_and_date' and 'amazon_fba_inventory_health' referenced in the workflow section.
- Boundary markers: None explicit; instructions rely on structured field processing.
- Capability inventory: Capabilities are limited to data extraction and joining (exports_create, sellers_and_vendors_list) with no write access or system shell access.
- Sanitization: Not explicitly defined, however, the skill processes specific numeric and categorical fields (SKU, price, percentage) which reduces the risk of interpreting embedded instructions.
Audit Metadata