Chou Fei

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent resource-ingestion and summarization helper, with some broad wording but no evidence of hidden data theft, persistence, or destructive behavior.

Install only if you want a skill that may be used for resource fetching and content processing. Avoid giving it private URLs, credential-bearing API targets, or sensitive files unless you have reviewed how your host agent will execute the skill and handle fetched content.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The fetch capability explicitly supports obtaining external resources but provides no warning about network access, remote content risks, or possible data-handling side effects. Users or orchestrators may invoke it without understanding that it can contact external systems, retrieve sensitive material, or process untrusted content.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill's stated responsibilities are broad—external resource acquisition, information ingestion, and compute allocation—without meaningful activation constraints or policy limits. In context, this makes the skill more dangerous because it combines data access and resource control, increasing the chance of overreach, abuse, or unsafe invocation against arbitrary targets.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal