Back to skill

Security audit

headache-relief-asmr

Security checks for vulnerabilities and agentic risk

Overview

This skill is not overtly malicious, but it should be reviewed because it can ask for binary gender in a headache-related interaction before giving simple ASMR links.

Before installing, consider whether users should be asked to disclose male/female gender for headache-relief audio recommendations. The skill should ideally make ASMR support opt-in, allow users to decline gender disclosure, and keep medical escalation guidance for severe or persistent symptoms.

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
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)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs the assistant to require a binary male/female classification before providing a recommendation, despite no clear safety or functional justification in the skill content. This creates unnecessary collection of sensitive personal data, excludes non-binary users, and can pressure users to disclose demographic information in a vulnerable health-related interaction.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation conditions are broad enough to trigger on ordinary mentions of headaches, migraines, or head pain, which can cause the skill to engage without a clear user request for this specific ASMR workflow. In a health-adjacent context, unintended activation increases the chance of irrelevant or inappropriate recommendations and may steer users into a scripted path that collects sensitive information unnecessarily.

Content

No source excerpt is available for this finding.