Nature-Style Academic Writer

Security checks across malware telemetry and agentic risk

Overview

This is a text-only academic editing skill with no code or system access; the main caveat is that it may broadly apply Nature-style editing and translate multilingual text into English.

Safe to install as a writing aid. Review outputs carefully for scientific accuracy, especially translated passages, and avoid submitting confidential unpublished material unless you are comfortable having your agent process it.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
88% confidence
Finding
The trigger phrases are broad enough to match ordinary writing-assistance requests, which can cause the skill to activate outside its intended scope. Overbroad activation is dangerous because it can unexpectedly steer unrelated user content into a specialized rewriting workflow, increasing the risk of unwanted transformation of sensitive, regulated, or user-intended text.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The skill instructs automatic translation of multilingual input into English without first obtaining user consent or confirming the target language. This is risky because translation is a substantive transformation that can alter meaning, discard source-language nuance, and violate user expectations or policy requirements when the user wanted editing in the original language.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal