li-xiaolai-perspective
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown files providing personality guidelines, decision-making workflows, and cognitive models. There are no executable scripts, binary files, or active code components in the provided content.
- [EXTERNAL_DOWNLOADS]: The skill contains numerous links to the author's official homepage (lixiaolai.com), GitHub repositories (github.com/xiaolai), and legitimate e-commerce sites (JD.com). These are documented as reference sources for the personality model and are well-known, safe domains.
- [PROMPT_INJECTION]: The skill includes instructions to guide the agent's behavior (e.g., 'Do not give specific buy/sell advice'), but these are standard role-playing constraints and do not attempt to bypass safety filters or override the system prompt in a malicious way.
- [DATA_EXFILTRATION]: No network-enabled tools or commands for data exfiltration were found. The 'allowed-tools' or specific API integrations are absent, and the skill relies on manual research logic.
- [NO_CODE]: There are no implementation files (Python, JavaScript, shell scripts) included in the provided analysis package, despite references to a 'scripts/' directory in the README. The analyzed content is purely instructional data.
Audit Metadata