Humanizer Academic

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent academic prose rewriting and detection aid, with no evidence of hidden data access, persistence, exfiltration, or destructive behavior.

Install only if you want an editor-style skill that rewrites academic prose and optionally runs a local AI-signal detector. Review outputs for factual fidelity and academic integrity, especially because the skill is explicitly designed to reduce AI-writing signals.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (1)

Description-Behavior Mismatch

High

Confidence: 97% confidence
Finding: The file content is a full Hong Kong economic report rather than an implementation, prompt, or evaluation artifact for the declared academic-humanizer skill. This creates a skill-definition/integrity mismatch: the agent may load or evaluate unrelated political/policy content instead of the intended transformation behavior, which can cause misrouting, unsafe outputs, or hidden agenda injection through substituted content.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal