ClawHub
Back to skill

Security audit

unisound-vaccination-reminder

Security checks across malware telemetry and agentic risk

Overview

Review recommended because this skill is purpose-aligned but can send raw resident health records to a remote model despite claiming de-identification and without an explicit consent step.

Install only if your organization has approved the remote model endpoint for resident health information. Do not include names, ID numbers, phone numbers, addresses, or other identifiers in the input unless the skill is updated to actually redact them and obtain explicit consent before sending data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill advertises handling sensitive vaccination and health-status inputs, and its documentation shows file read, file write, and network capabilities, but it declares no permissions or explicit disclosure of those capabilities in a machine-enforceable way. This creates a trust and governance gap: users or platforms may approve the skill without understanding that protected health-related data can be read from disk, transmitted to a remote API, and written to output files.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The stated purpose suggests a local reminder-generation skill, but the behavior includes sending residents' vaccination records and special health conditions to an external network service using a provided appkey. In a medical/public-health context, this is especially dangerous because the transmitted data is health information, and the mismatch can prevent informed consent, proper compliance review, and appropriate safeguards for third-party processing.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill sends resident vaccination history and health conditions to a remote LLM API, which constitutes transmission of sensitive medical data to an external service. There is no explicit consent flow, privacy notice, minimization step, or guarantee that the user understands where the data is being sent; in a healthcare context this materially increases privacy, compliance, and data-governance risk.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.