product-selection
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of educational documentation and procedural instructions. It contains no executable scripts, automated network calls, or commands that modify the system environment.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a process that involves the agent analyzing external content such as Xiaohongshu search results, trending keywords, and data exported from third-party analytics platforms (like Qiangua). This establishes a potential surface for indirect prompt injection.
- Ingestion points: Data entering the agent's context through search result analysis and imported Excel/data files.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' guards when the agent processes this external data.
- Capability inventory: The skill defines no automated tools or system-level capabilities; the analysis is restricted to logical reasoning and data interpretation within the conversation.
- Sanitization: No sanitization steps are defined for handling user-provided or platform-scraped research data.
Audit Metadata