Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent project risk assessment skill with no evidence of hidden execution, persistence, credential handling, or data exfiltration.
Before installing, consider that this skill is designed for Chinese-language招商 and project landing review. It may use web and enterprise-credit style checks, so provide only project and company information needed for the assessment and verify important legal, financial, policy, and credit conclusions against authoritative sources.
The display name, summary, triggers, and all operating instructions are written exclusively in Chinese, and the skill does not state that output language follows user preference or that the locale restriction is intentional. Under the policy, forcing a specific language without opt-in is a natural-language policy concern.
The trigger list is broad and overlaps with generic project analysis or risk-review requests, which can cause unintended activation outside the intended招商/项目落地风控 scope. In an agent environment, over-broad invocation can route unrelated user queries into a high-authority workflow that may produce inappropriate assessments, collect unnecessary data, or crowd out a more suitable skill.
No suspicious patterns detected.
