product-categorization
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of untrusted external data (product titles and descriptions from supplier catalogs) via AI for automatic categorization. This creates an attack surface where malicious instructions embedded in product data could attempt to influence the agent's behavior.
- Ingestion points: External supplier catalogs (SKILL.md) and input JSON files (task.md).
- Boundary markers: No specific delimiters or 'ignore' instructions are mandated in the prompt templates shown.
- Capability inventory: The skill's suggested logic involves database writes and JSON generation based on processed data.
- Sanitization: The skill recommends a human-review workflow to validate AI suggestions, which significantly reduces the risk of incorrect categorization being published.
Audit Metadata