Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed research-paper writing aid with no bundled executable code, persistence, credential access, or hidden security-impacting behavior.
Installers should treat this as a paper-drafting/template helper and specify the desired output language and citation style when using it. The referenced Python command is not included in the package, so expect the skill to function as instructions rather than as a runnable tool unless another component supplies that script.
The skill content is primarily written in Chinese and does not offer users a language choice or document a locale restriction. In a general-purpose research paper writing skill, this can cause unintended language steering, producing outputs in the wrong language and creating reliability, usability, and policy-compliance issues for users expecting English or multilingual support.
The trigger keywords are broad and include generic phrases like 'research paper', 'generate paper', and 'thesis writing', which can cause the skill to activate for ordinary writing-assistance requests outside the user's intent. This creates unintended invocation risk, where a high-scope content-generation skill may intercept unrelated prompts and produce misleading or undesired outputs.
The manifest mixes a Chinese-only description with both Chinese and English trigger keywords, but it does not state whether users can choose their preferred output language. Because natural-language policy review applies to all file types, this can be a locale/language policy concern if the skill defaults language behavior without explicit user choice.
No suspicious patterns detected.
