Security audit

Sleep Quality Coach

Security checks for vulnerabilities and agentic risk

Overview

This is a non-executable sleep coaching skill with coherent behavior, but users should treat its supplement and diagnostic language as general education, not medical advice.

Reasonable to install for sleep-habit coaching, routines, and tracking prompts. Do not rely on it to diagnose a condition or decide on supplements or medications; consult a qualified clinician for persistent insomnia, severe sleepiness, breathing pauses, restless legs symptoms, significant anxiety or depression, pregnancy, chronic illness, or medication interactions.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (3)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The skill gives actionable supplement guidance, including melatonin for circadian rhythm, jet lag, and shift work, plus magnesium, lavender, and valerian-related advice, without placing a clear safety warning directly with those instructions. In a health skill, users may interpret this as personalized treatment advice despite contraindications, interactions, dosing uncertainty, and special risks for pregnancy, chronic illness, psychiatric conditions, liver issues, or concurrent sedative use.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill claims it does not provide medical diagnosis, but earlier sections repeatedly instruct the agent to identify sleep problem types, detect red flags, and name specific suspected conditions such as sleep apnea, RLS, circadian rhythm disorder, and anxiety-driven insomnia. This contradiction can cause users to rely on quasi-diagnostic outputs as medical guidance, especially in a health context where users may defer professional care or act on incorrect classifications.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 71% confidence
Finding: Natural-language examples switch into Chinese for a dedicated section, but the document does not explicitly say the skill adapts to the user's chosen language. This can be read as a locale/language assumption rather than an opt-in choice, which risks a language policy issue.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.