ClawHub
Back to skill

Security audit

Network Tox Docking Research Planner

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only biomedical research-planning skill with no code execution, credential use, persistence, or hidden data access.

Installers should treat this as a research-planning aid, not a validated biomedical authority. Review outputs for scientific accuracy, avoid overclaiming causality or docking evidence, and use appropriate expert review before relying on plans for publication or health-related decisions.

SkillSpector

By NVIDIA
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)

Vague Triggers

High
Confidence
93% confidence
Finding
The trigger definition is excessively broad, including casual phrases like wanting to study how a chemical affects a disease, which can cause the skill to activate outside clearly intended contexts. This increases the chance of misrouting user requests, bypassing more appropriate tools or safeguards, and producing research-planning output when the user did not explicitly request this specialized workflow.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The document instructs users to activate a recovery sequence whenever 'Hard Rule 11' fires, but no such rule is defined in this file or surrounding context. This creates an ambiguous control condition that can lead to inconsistent execution, accidental omission of validation steps, or arbitrary interpretation by an agent or operator.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal