ClawHub
Back to skill

Security audit

Cross Domain Engine

Security checks across malware telemetry and agentic risk

Overview

This skill is a text-only research prompt for cross-domain correlation, with broad high-stakes use cases but no executable code, hidden access, persistence mechanism, or data exfiltration behavior.

Install only if you want a brainstorming framework for cross-domain research. Do not rely on it by itself for investment, security, legal, patent, or enforcement decisions; require corroborating sources and human review before acting on any generated correlation.

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

Medium
Confidence
91% confidence
Finding
The skill description defines very broad usage conditions such as discovering correlations, threat cross-referencing, investment signals, and innovation mining without clear scope boundaries or disallowed contexts. In an agent environment, this can cause over-invocation or invocation in sensitive decision-making workflows, increasing the chance that speculative or weakly validated outputs influence security, financial, or research actions.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The "When to Use" section encourages activation across a wide range of loosely defined scenarios and provides no negative examples or stopping conditions. This makes the skill easier for an orchestrating agent to invoke inappropriately, especially for high-impact tasks like threat analysis, investment signal generation, and patent-mining where speculative correlations may be mistaken for actionable evidence.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.