Security audit
Security checks for vulnerabilities and agentic risk
This skill is a documentation-only mapping template for organizing multi-agent work and does not contain executable, hidden, or data-accessing behavior.
Installers should expect a Chinese-language planning template that may encourage structured configuration output for multi-agent workflows. Review or narrow the trigger description if you do not want mentions of GitHub or data analysis to activate this skill outside the 三省六部 mapping context.
The publishing note says the skill description should cover trigger terms such as “技能匹配”, “GitHub”, and “数据分析”, several of which are broad everyday/workplace concepts rather than narrowly scoped invocation phrases. Because this is a markdown file and the line provides trigger guidance without negative examples or explicit scope limits, it could lead to overly broad or ambiguous skill activation.
This manifest contains user-facing natural-language descriptions exclusively in Chinese across the title, description, role labels, responsibilities, and workflow steps. Under the policy, forcing a specific language without user opt-in or a clearly documented locale-specific justification is a natural-language policy violation.
No suspicious patterns detected.
