Medical Medium Liver Rescue

• Outcome: Review
• Latest audit: Jun 9, 2026 at 10:51 AM UTC
• Version:
  1.0.0

Overview

This text-only health skill has no code or credential behavior, but it gives authoritative alternative medical and cleanse guidance for real conditions without adequate runtime safety boundaries.

Install only if you deliberately want an alternative-health book-summary skill and will not treat it as medical advice. Do not use it for diagnosis, urgent symptoms, medication decisions, pregnancy, diabetes, liver or gallbladder disease, eating-disorder risk, or chronic-condition management without a licensed clinician.

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 (6)

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The trigger list is extremely broad and includes common medical and wellness terms such as liver, skin, bloating, constipation, acne, autoimmune, and detox. This can cause the skill to activate in many ordinary health discussions and inject unverified Medical Medium claims into unrelated or clinically sensitive contexts, increasing the chance of harmful misinformation displacing evidence-based guidance.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: The skill presents cleanse protocols, disease-causation claims, and dietary recommendations as authoritative medical guidance without any warning that these claims are not medically validated or that users should consult a licensed clinician. In a health context, especially where the content attributes eczema, psoriasis, SIBO, fatty liver, EBV, and autoimmune illness to unsupported mechanisms, this creates a meaningful risk of delayed care, unsafe self-treatment, and harmful dietary restriction.

Natural-Language Policy Violations

High

• Confidence: 99% confidence
• Finding: The file presents numerous unsupported and medically dubious claims as established fact, including assertions that common skin diseases are fundamentally liver conditions, that standard tests miss severe liver dysfunction, and that specific foods or cleanses can reverse chronic illness. In a health-related skill explicitly triggered by users seeking help for fatty liver, SIBO, eczema, autoimmune disease, and detox protocols, this can mislead users away from evidence-based diagnosis and treatment and may cause harmful self-management or delayed medical care.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This content gives prescriptive cleanse and elimination-diet instructions for liver-related and chronic health conditions while presenting unsupported medical claims as safe and effective, without advising users to seek clinician guidance or consider contraindications. In a health skill, users may treat this as actionable medical advice, which can delay proper diagnosis, worsen underlying disease, or cause harm in people with diabetes, eating disorders, pregnancy, liver/gallbladder disease, or medication-dependent conditions.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: This file provides explicit health and treatment guidance, including claims about causes of disease, dismissal of standard testing, discouragement of common medications, and endorsement of specific detox protocols, without any disclaimer or direction to seek licensed medical care. In the context of a medical-adjacent skill centered on unverified Medical Medium claims, the absence of a warning materially increases the risk that users will rely on unsafe or misleading advice for real health decisions.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The file explicitly frames unverified medical claims as authoritative guidance for chronic conditions and instructs the agent to avoid defensiveness about the purported supernatural source. In a health context, presenting alternative-treatment claims without a clear medical disclaimer or safety boundary can mislead vulnerable users into delaying evidence-based care or following unsafe detox/cleanse advice.

VirusTotal

VirusTotal findings are pending for this skill version.