Scientific Thinking

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk scientific reasoning skill; its main caveat is that it can be invoked automatically and may shape answers in research-adjacent contexts.

Install this if you want a structured scientific reasoning style available to your agent. Consider disabling or narrowing implicit invocation if you only want it used on explicit research, experiment-design, or evidence-evaluation requests, especially around medical, safety-critical, or other high-stakes topics.

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: 93% confidence
Finding: The skill enables allow_implicit_invocation without any visible trigger constraints, scoping, or exclusion conditions, so the system may auto-select this skill in contexts the author did not intend. Because this skill influences reasoning and interpretation rather than performing a narrow bounded action, over-broad implicit activation can silently shape scientific conclusions, evidence framing, or experiment suggestions in inappropriate contexts.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal