Security audit
Security checks for vulnerabilities and agentic risk
This Chinese question-coaching skill is a Markdown-only instructional skill; its broad triggers and Chinese-only output are usability concerns, not evidence of unsafe access or malicious behavior.
Install this if you want a Chinese-language assistant for practicing deeper questioning. Be aware that generic phrases like asking for help with questions or deep thinking may activate it, and only enable any scheduled practice workflow if you actually want recurring prompts.
The trigger list contains broad, everyday phrases such as '帮我提问', '如何提问', and '深度思考', which can match many normal conversations outside the skill’s intended scope. Overbroad activators can cause unintended routing into this skill, leading to prompt/intent hijacking at the orchestration layer and responses that override user expectations or conflict with more appropriate tools.
The rule '用户描述内容即自动进入实战模式' is an ambiguous auto-activation condition that can fire on almost any user message describing something they read, heard, or learned. This increases the chance of unsolicited skill takeover, where the assistant shifts into a rigid questioning workflow without clear consent, potentially suppressing the user’s actual task or causing cross-skill misrouting.
Mandating Chinese output without user opt-in can override the user’s language preference and platform defaults, which is a control-plane safety issue even if not directly exploitable for code execution or data theft. In multilingual environments, this can degrade usability, hide important safety messaging from the user, and create confusion if the original request was in another language.
The file title and all instructional content are written exclusively in Chinese, with no natural-language indication that users may choose another language or locale. Under the policy rule for language/locale constraints, this can be a violation when a skill implicitly requires a specific language without user opt-in or justification.
No suspicious patterns detected.
