Security audit
Security checks for vulnerabilities and agentic risk
This is a local writing-style helper with no executable code, network behavior, credential handling, or persistence, though its activation wording is broader than ideal.
Before installing, consider narrowing the trigger wording so the skill activates only when Wang Xiaobo-style writing, analysis, or named techniques are explicitly requested. No security-sensitive behavior was found in the inspected artifacts.
The trigger list includes broad terms such as 科普写作, 趣味写作, logical deduction writing, and humorous prose, which can match many ordinary writing requests unrelated to Wang Xiaobo-style transformation. Over-broad activation can cause the agent to invoke this skill unexpectedly, steering outputs toward a specific stylistic framing and reference-loading behavior when the user did not clearly ask for it.
The scope states that the skill works for any Chinese-language writing task involving engaging, logical, and humorous prose, which is broad enough to overlap with a large share of normal writing assistance requests. This creates prompt-squatting risk: the skill may activate in situations where the user wants generic writing help, causing unintended style imitation and unnecessary dependence on this skill's instructions.
No suspicious patterns detected.
