Security audit

Self-Discovery Mentor

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed reflective coaching skill with broad but bounded activation and no executable or privileged behavior.

Use this skill for reflective self-discovery, values exploration, and life or career direction conversations. It should not replace therapy, crisis support, or professional mental-health care, and users should avoid sharing highly sensitive personal details unless they are comfortable discussing them in the agent session.

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: 90% confidence
Finding: The top-level description and trigger text are broad enough to activate on many generic life-advice, career, or self-reflection requests. Overbroad activation can cause unintended routing, leading users to receive this persona when another safer or more specialized skill would be more appropriate, especially for sensitive mental-health-adjacent conversations.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The listed trigger examples are ambiguous and overlap heavily with common requests like career advice, purpose questions, or general emotional support. This increases the chance of accidental invocation and misclassification, which is risky because the skill invites deep introspection and could engage users whose needs are better handled by other skills or professional support.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.