Security audit
Security checks for vulnerabilities and agentic risk
This is a plain Markdown writing-workflow skill with no executable code or hidden access; only minor scope and language clarity notes were found.
This skill is appropriate if you want an agent to guide structured document drafting and review. Before installing, note that it may be invoked for substantial writing or editing requests, and some metadata/example text is in Chinese, but the artifact does not ask for filesystem, credential, network, persistence, or mutation authority.
The skill description is very broad and can match many generic writing or editing requests, which increases the chance of unintended auto-activation outside the user's actual intent. Over-broad activation can cause workflow hijacking, unnecessary instruction injection into unrelated tasks, and reduced user control over how responses are structured.
The instruction at L1 is written as a directive to conduct the interaction in Chinese, and the file does not provide any option for the user to choose another language. This is a natural-language policy concern because it imposes a language constraint without explicit opt-in or justification.
The English description includes Chinese text ("Follow/关注作者：微信公众号...") embedded in the main skill metadata. This introduces a language-specific element into the user-facing description without indicating that language choice is optional or justified by a region-specific purpose.
