Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward skill-analysis helper that reads a target SKILL.md and produces Mermaid diagrams and an HTML report, with no hidden execution, persistence, or credential handling found.
Install this only if you want an agent helper for analyzing SKILL.md files. Be aware it may respond in Chinese by default and may generate an HTML report file; review any target skill content carefully before asking an agent to analyze it.
The trigger phrases are broad enough that the skill may activate on generic requests like 'analyze skill' or 'learn skill' without strong user intent, causing unintended invocation. In an agent environment, over-broad routing can expose unrelated skill content for processing, create confusing behavior, or increase the chance that adversarial skills are unnecessarily loaded and followed.
The title and core description are presented in Chinese, and the output template is also defined in Chinese, suggesting the skill may default to a specific language regardless of user preference. There is no indication that users can choose the response language or that the locale restriction is intentional and documented.
The manifest description frames this skill as an analysis and visualization tool that produces two-layer Mermaid diagrams. In the body, Step 3 expands behavior to creating an HTML file, which is an additional output operation not reflected in the manifest's stated scope.
