Security audit

Ashin Perspective

Security checks across malware telemetry and agentic risk

Overview

This is a text-only Chinese persona skill for simulating Ashin’s public style, with no code execution, credential access, network use, or hidden install behavior.

Install this only if you want a Chinese-language AI roleplay of Ashin’s public persona. Treat responses as stylized AI-generated perspective, not statements from the real person, and use the documented exit phrases if the persona activates when you do not want it.

SkillSpector

By NVIDIA

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: 91% confidence
Finding: The trigger phrases are intentionally broad and include common conversational wording such as asking to use someone's perspective or to switch modes. That can cause the skill to activate unexpectedly and override the assistant's normal behavior, increasing the chance of persona hijacking, misrouting, or bypass of more appropriate skills and safety-aware default responses.

Natural-Language Policy Violations

Medium

Confidence: 78% confidence
Finding: The skill mandates a fixed persona, tone, and first-person identity without checking user preference or preserving normal assistant transparency. In this context, that is risky because it encourages identity simulation and suppresses clarifying meta-communication, which can mislead users about authorship or reduce the model's willingness to state uncertainty plainly.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.