Security audit

Cao De Wang Perspective V2

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only business-perspective persona skill with no code execution, credential access, persistence, or hidden system changes.

Install this only if you want a stylized business decision lens. Treat its output as interpretive guidance, not as factual statements from Cao Dewang or as legal, financial, regulatory, or investment advice; invoke it explicitly to avoid accidental persona-style responses.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger list includes broad business terms such as '专注主义' and '敬天爱人' that can plausibly appear in normal discussion without an explicit request to invoke the persona. That can cause unintended activation of the skill, leading the assistant to switch into a fixed perspective when the user only meant to discuss those concepts, but the content itself does not enable code execution, privilege escalation, or direct data exfiltration.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The activation section says the skill can be invoked by several natural-language phrases but does not define firm boundaries for activation versus ordinary mention. In practice, that ambiguity increases the chance of accidental persona switching and response steering, especially in business or strategy conversations where similar wording is common.

Static analysis

No suspicious patterns detected.