The Greatest Secret

• Outcome: Review
• Latest audit: Jun 15, 2026 at 6:09 AM UTC
• Version:
  1.0.0

Overview

This is a text-only spiritual self-help skill, but it is too willing to respond to serious depression, anxiety, and hopelessness without clear mental-health safety boundaries.

Review carefully before installing. This skill appears non-malicious and technically low risk, but users should treat it as spiritual/self-help content only, not mental-health care. It should not be relied on for persistent depression, severe anxiety, hopelessness, self-harm thoughts, medication or therapy decisions, or any situation where a licensed professional or crisis resource is appropriate.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The trigger phrases are broad, common distress statements like 'I'm stressed' or 'I feel lost,' which can cause the skill to activate in many situations unrelated to this book or appropriate self-help guidance. In a mental-health-adjacent context, this increases the chance of the assistant steering vulnerable users into a rigid spiritual framework instead of giving neutral support or safer escalation.

Vague Triggers

Medium

• Confidence: 82% confidence
• Finding: The Quick Start says the AI should appear 'whenever I sense this book could help' and must proactively present onboarding, which relies on subjective judgment rather than clear consent or scoped invocation. That ambiguity can lead to unsolicited intervention during sensitive conversations, especially around anxiety, identity, or suffering, where an imposed spiritual framing may be inappropriate.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The skill explicitly targets anxiety, stress, negative feelings, and suffering, but provides no warning that it is not mental health care and no boundary for crisis situations. Because it presents authoritative guidance in response to distress, users may rely on it instead of seeking qualified support, increasing risk of delayed care or harmful minimization of serious symptoms.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: The invocation test includes a user with months of depression and anxiety, failed therapy and medication, and statements about running out of hope, but instructs only spiritual reframing and introspection without any safety check or crisis guidance. This is especially dangerous because hopelessness is a known acute risk indicator, and the skill normalizes bypassing clinical escalation in favor of metaphysical advice.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This content gives self-help guidance for fear, anxiety, anger, sadness, and emotional suffering while framing those states as consequences of mistaken beliefs and suggesting users can resolve them by changing their relationship to thoughts. In a skill explicitly triggered by users reporting stress, anxiety, feeling lost, or suffering, the lack of a clear boundary that this is not mental health care may discourage appropriate professional support or oversimplify serious conditions.

VirusTotal

63/63 vendors flagged this skill as clean.