Security audit
Security checks for vulnerabilities and agentic risk
This is a simple analogy-formatting skill with no code execution, persistence, credential access, or hidden data handling.
Installers should know this skill may answer in a Chinese-style template and may activate for broad requests about more examples or different angles. It does not show signs of accessing private data, running commands, or changing the environment.
The documentation is entirely in Chinese and presents the output format and examples in Chinese without indicating language adaptation to the user's preference. In a multi-language agent environment, this can cause the skill to respond in an unintended language, degrading usability and potentially interfering with downstream workflows that expect the user's original language.
The trigger phrases are broad enough to match normal conversational requests such as asking for more examples or another explanation angle, which can cause the skill to activate when the user did not specifically want this formatting. In an agent system, over-broad activation can hijack routing and override more appropriate skills, reducing reliability and potentially causing unintended behavior.
No suspicious patterns detected.
