Security audit

paper-aigc-reducer

Security checks across malware telemetry and agentic risk

Overview

This skill is a text-rewriting helper whose disclosed purpose is to reduce AI-detection signals in Chinese academic or technical writing, so users should review it carefully before use.

Install only if you intentionally want an AI-detection-evasion rewriting tool. It does not appear to access your files or run code, but its purpose may be inappropriate for academic integrity, platform rules, or disclosure requirements; ordinary proofreading or clarity editing should use a narrower writing-assistance skill instead.

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: 87% confidence
Finding: 技能触发描述使用了“降AIGC、降重、论文改写、去除AI痕迹、让文本更像人写的、通过AI检测”等宽泛且高度语义化的短语，容易在相邻但不应触发的写作、润色或编辑场景中被误调用。由于该技能本身面向规避检测，误触发会把普通写作请求导向规避性改写，从而放大合规与滥用风险。

Ssd 2

Medium

Confidence: 98% confidence
Finding: 技能描述明确围绕“降低AIGC检测率”“去除AI痕迹”“通过AI检测”等目标设计，实质上是在帮助用户规避内容来源检测与审查机制。这类能力可被用于学术不端、欺骗性内容伪装和绕过平台治理，且全文规则都在系统化优化规避效果，使危险性因上下文而显著升高。

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.