Security audit

Academic Writing Refiner

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only academic writing helper with no evidence of unsafe commands, credential use, persistence, or hidden data movement.

Safe to install for academic manuscript editing. Be aware it may activate for a wide range of research-writing polish requests, so use it when you want that style of assistance and avoid sharing confidential unpublished manuscript text unless you are comfortable processing it with your agent.

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: 93% confidence
Finding: The trigger criteria are very broad and include generic requests such as 'improve', 'edit', 'proofread', or any pasted LaTeX related to writing help. This can cause the skill to activate for many ordinary writing tasks outside its intended academic-paper scope, increasing the chance of incorrect routing, unnecessary exposure of user content to the skill, and degraded downstream behavior.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger description is very broad, covering generic requests like improve, polish, edit, or proofread and many common academic-writing contexts. This can cause the skill to activate when a more specific or safer skill would be more appropriate, leading to incorrect routing, unnecessary exposure of user content to this skill, or degraded system behavior through prompt over-capture.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.