ClawHub

LLM数据自动化

Security checks across malware telemetry and agentic risk

Overview

This skill is coherent for data automation, but it asks users to run generated file-processing code and use a third-party AI API without enough safety or privacy disclosure.

Review generated code before running it, work on copies of important data, keep outputs in a separate directory, and avoid scheduled runs until paths and overwrite behavior are clear. Use local Ollama for private or regulated documents when possible; only use the recommended third-party API after confirming what data is transmitted, retained, and allowed under your organization’s policies.

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 (2)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly promotes LLM-generated Python code that users can run directly against local Excel/CSV/PDF files, but it does not warn that generated code may overwrite files, leak sensitive data, or perform unintended transformations. In this context, the danger is elevated because the whole value proposition is zero-code execution by non-experts, who may trust and run file-system-affecting code without reviewing it.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill recommends configuring a third-party cloud API key and using an external API relay service, but it does not disclose that prompts or attached data may be transmitted to outside providers. Because this skill is designed for business documents, spreadsheets, PDFs, and multi-source data integration, users may unknowingly send sensitive or regulated data to external services.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal