Security audit
Security checks for vulnerabilities and agentic risk
This skill is a simple document summarizer with no code execution, network access, persistence, or credential handling.
Install this if you want a lightweight technical document summarizer. Be aware that generic requests like "summarize" may activate it, and its default output is Markdown.
The invocation phrases are broad and generic, especially terms like 'summarize', 'summary', and 'extract key points', which can match many unrelated user requests. This can cause the skill to activate unintentionally, overriding user expectations or applying document-processing behavior in contexts where it was not explicitly requested.
The invocation phrases are generic and likely to match ordinary user requests such as 'summarize' or 'summary,' which can cause the skill to activate outside narrow intended contexts. Overbroad activation increases the chance of unintended routing, priority hijacking, or accidental application of this skill to sensitive content the user did not explicitly mean to process with it.
The file contains prescriptive natural-language instructions to always produce output in Markdown and the surrounding metadata and headings are in Chinese, but there is no explicit language-choice guidance for users. While the stronger issue here is formatting rather than locale, the document does not clearly offer user choice on language or locale behavior.
The instructions require the model to "Produce output in Markdown format" and "Return valid Markdown only" as a fixed behavior. This enforces a single output format without any user opt-in or documented justification, which may conflict with organizational policy favoring user choice in presentation or locale-related preferences.
No suspicious patterns detected.
