Cybersec Helper

Security checks across malware telemetry and agentic risk

Overview

This is a text-only cybersecurity guidance skill that is broadly activated but clearly scoped toward ethical, authorized security review and does not add hidden access or execution behavior.

Install this if you want security-review guidance to appear readily in security-related conversations. Before using it for testing, keep targets explicitly authorized and scoped. The reviewed artifact does not show code execution, credential handling, hidden data access, destructive behavior, or persistence.

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 (1)

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The trigger conditions are broad enough that ordinary mentions of security-related topics such as 'is this safe?' or 'vulnerabilities' can activate the skill in many unrelated contexts. Because this skill changes model behavior and provides security-oriented guidance, unintended activation could cause overreach, unnecessary handling of sensitive security topics, or policy-friction in benign conversations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal