ClawHub
Back to skill

Security audit

企业资金健康度评估

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only business risk report skill that uses public information and does not show hidden execution, credential access, or data exfiltration.

Before installing, understand that this skill produces public-information financial risk assessments about companies. Verify citations and source freshness, especially for non-public companies with limited financial disclosure, and treat any generated Word document as a persistent local file that may contain sensitive analysis.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
Instructing the agent to deliver the full report as a Word document can trigger file-generation behavior or external artifact creation without explicit user consent or a warning about side effects. In agent environments with document-writing capabilities, this may lead to unexpected local file creation, persistence of sensitive analysis content, or accidental transmission through connected storage/workflow systems.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.