Security audit

Cognitive Load Theory

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only instructional-design skill with no executable code, privileged access, or hidden data handling.

Safe to install for CLT-based help with training, onboarding, documentation, and learning-material redesign. Be aware it may activate on broad phrases like “this is overwhelming,” and do not persist real user or company examples into the skill unless that is intended.

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: 92% confidence
Finding: The manifest-style description says to activate when someone says phrases such as 'this is too much to take in at once,' 'how do I design training that actually works,' and 'this documentation is hard to follow.' These are broad, everyday statements that could appear in many contexts, and the file does not define a narrow invocation boundary beyond a couple of exclusions, increasing the risk of unintended activation.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: Phrases like 'Designing training, documentation, onboarding,' 'Learners aren't understanding,' and 'this is overwhelming' are conceptually relevant but still broad enough to match common workplace discussion. Although one 'Not when' condition is given, the file lacks concrete negative examples or tighter boundaries distinguishing this skill from general coaching, UX, or documentation advice.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.