Security audit
Security checks for vulnerabilities and agentic risk
This is a writing-style guidance skill with no code execution, credential access, persistence, or hidden data handling.
Installers should understand that this skill may activate for broad requests about logical Chinese prose and steer responses toward Wang Xiaobo-style writing; review that scope if you only want it triggered by explicit style requests.
The description frames the assistant as teaching Wang Xiaobo essay writing specifically "in Chinese writing," which implies a fixed language behavior. There is no accompanying statement that the user may choose another language or explicitly opt into Chinese-only output.
The phrase "or wants logical, engaging Chinese prose" is not specific to this skill and overlaps with common writing-assistance requests that may not intend Wang Xiaobo-style imitation. The file does not provide clear scope boundaries or negative examples to distinguish general Chinese writing help from explicit style invocation.
This markdown file presents the analysis and transformation example entirely in English while referencing Chinese source material and a named Chinese authorial style. Under the policy criteria, this can be read as imposing a specific language/locale presentation without any opt-in or alternative language option.
No suspicious patterns detected.
