Values Clarification

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable guided reflection skill, with only minor cautions around sensitive personal notes and opt-in use during difficult life events.

Safe to install for guided self-reflection. Treat outputs as sensitive personal data, avoid using it as a substitute for professional or emergency support, and only grant filesystem access if you deliberately want the agent to save or revisit your notes.

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: 77% confidence
Finding: The trigger condition `user_seeking_values_work` is undefined and lacks clear scoping, so the automation could activate on vague or misclassified user signals. In a mental-health-adjacent skill, unsolicited initiation of values work after crisis or dissatisfaction cues can steer vulnerable users into a structured intervention without adequate consent, suitability checks, or escalation boundaries.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal