qiye-risk-platform
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a step-by-step restoration manual for creating a risk management platform. All analyzed instructions are related to UI generation and legitimate business logic based on the vendor's (Patsnap) domain.
- [DATA_EXPOSURE_EXFILTRATION]: The skill references data from 'open.zhihuiya.com' and 'patsnap.com'. These are official domains associated with the author 'patsnap'. There are no instances of credential harvesting or exfiltration to third-party domains.
- [PROMPT_INJECTION]: While the skill contains direct instructions for the AI agent (prompts for the user to copy-paste), these are functional requests to build a specific tool. No instructions attempt to bypass safety filters or override the agent's core constraints.
- [INDIRECT_PROMPT_INJECTION]: The generated platform accepts user input (enterprise names) to generate reports, which is a common attack surface for indirect injection. However, as this is a construction manual, the risk is architectural and handled by the agent's code generation safety, placing it at a low/safe level.
- [REMOTE_CODE_EXECUTION]: No remote scripts are downloaded or executed. The skill purely generates static HTML and local logic within the agent environment.
Audit Metadata