Intuition

Security checks across malware telemetry and agentic risk

Overview

This documentation-only skill changes the agent toward fast pattern-based answers, with disclosed limits and no code execution or hidden system access.

Install this only if you want quicker, more decisive first impressions. For medical, legal, financial, security, or major business decisions, explicitly ask the agent to analyze carefully, compare options, explain uncertainty, and avoid storing calibration notes unless you know where those notes are kept.

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: Defaulting to intuitive mode when not explicitly specified can cause the skill to activate in ordinary conversations without clear user consent. In a safety-sensitive assistant, this may suppress deliberation and increase the chance of overly confident, low-justification responses in contexts where analytical reasoning is needed.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger 'Time pressure indicated' is ambiguous and can be inferred from many normal user messages, making activation subjective and overly broad. This creates a path for the assistant to prematurely switch into fast, confident response behavior even when the task is complex, unfamiliar, or safety-relevant.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal