Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Chinese-language business evaluation skill made only of markdown guidance, with no executable code, external access, persistence, or hidden data handling.
Install this only if you want a Chinese-language Gate0/Gate1 business-scenario scoring workflow. Treat its ratings as advisory because the trigger is broad and the evidence rules leave room for inference when hard evidence is missing.
The trigger condition is phrased broadly enough to activate on generic business-assessment or investment-review requests, which can cause the wrong skill to intercept unrelated user tasks. In an agent setting, over-broad routing increases the chance of misapplication, producing misleading evaluations or bypassing more appropriate domain-specific safeguards.
This markdown file is entirely written as prescriptive Chinese-language process guidance, including a required Chinese output template, but it does not indicate that language selection is optional or limited to a justified region-specific context. Under the policy rule for natural-language violations, forcing a specific language without user opt-in is in scope.
The rules state that scoring must rely on user-provided evidence and cap unsupported items, but also say missing evidence may be replaced by logical inference. In a gatekeeping workflow, this contradiction can be exploited to justify favorable or inconsistent evaluations, weakening the integrity and repeatability of decisions.
The Gate0 section says all four dimensions must pass and that missing or incomplete evidence causes immediate elimination, yet nearby text allows logical inference as a substitute. Because Gate0 is the hard filter for project approval, this ambiguity enables bypass of strategic red lines or arbitrary rejection depending on reviewer preference.
The file mandates Chinese-only instructions and output format throughout, without offering any user language choice or fallback. This can cause the agent to ignore or override a user's preferred language, reducing transparency and increasing the chance that users misunderstand security-relevant gating decisions or evidence requirements.
This markdown file contains user-facing instructions exclusively in Chinese, including headings, scoring rules, and output templates. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly documented and justified, which is not present here.
The file presents the skill’s description, operating instructions, and examples entirely in Chinese, but does not state that language selection is optional or user-chosen. Under the policy, forcing a specific language or locale without opt-in is a natural-language policy concern.
No suspicious patterns detected.
