Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Chinese humor personality-quiz skill with no code execution, persistence, credential access, or hidden data flow.
Install this if you want a Chinese, irreverent parody personality test that may start when SBTI or shared SBTI results are discussed. It does not appear to run code or access private files, but users should expect joking, mildly insulting personality-result language and Chinese-only output.
Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.
### 纯S — 海豚 (Dolphin)
代号: SS
社交界的天花板，跟谁都能聊到起飞。朋友遍天下，但有多少是真朋友不重要，重要的是热闹。特征：聚会的核心，never say no，手机通讯录2000+且都备注了emoji。
### 纯E — 水母 (Jellyfish)
代号: EE
Triggering when 'someone shared an SBTI result' is ambiguous because mention of an SBTI result does not necessarily mean the user wants to launch the skill. In context, this makes accidental activation more likely during casual discussion, which can interrupt the conversation flow and produce an unwanted quiz experience.
Mandating Chinese-only questions and results without considering the user's language preference can create accessibility and usability problems, especially in multilingual environments. While not a classic security flaw, it can cause the skill to behave unexpectedly for users who did not opt into Chinese output, reducing transparency and user control.
Triggering when 'someone shared an SBTI result' is ambiguous because mention of an SBTI result does not necessarily mean the user wants to launch the skill. In context, this makes accidental activation more likely during casual discussion, which can interrupt the conversation flow and produce an unwanted quiz experience.
No suspicious patterns detected.