ClawHub
Back to skill

Security audit

unisound-paper-writing-assist

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent medical writing helper, but it sends the user’s provided notes to a remote model API.

Install only if sending the provided notes to the disclosed remote API is acceptable for your organization. Do not include patient identifiers, regulated health information, or confidential unpublished study details unless you have approval for that endpoint and app key handling.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documentation advertises executable behavior that reads input files, writes output files, and sends content to a remote API via an app key, but it does not declare any permissions. Undeclared file and network capabilities reduce transparency and can bypass user or platform review expectations, especially because the uploaded notes may contain unpublished clinical research or sensitive data.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill sends user-provided notes, which in this medical research context may contain sensitive clinical or unpublished study content, to a remote third-party API endpoint without any explicit runtime disclosure, consent check, or data-classification safeguard. In a healthcare setting this materially increases confidentiality, compliance, and data-governance risk, especially if users assume processing is local or approved for regulated data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.