ClawHub

Data Generator

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a straightforward training-data generator, but users should choose the LLM endpoint carefully because command text may be sent there.

Before installing, confirm that API_URL or the provided base_url points to a model service you trust, avoid including secrets or private data in command text, and note that the documented Excel/custom-prompt examples do not match the current Python interface.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill declares only a binary requirement (`python3`) but the content and static analysis indicate capabilities related to environment access, file writing, and network activity without corresponding permission disclosure. This is dangerous because users and policy engines may assume the skill is low-privilege while it can read local inputs, write generated datasets, and potentially send data externally, creating a transparency and consent gap.

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill reads API credentials from environment variables and sends user-provided commands to an LLM endpoint, including arbitrary base URLs from configuration. This creates an undeclared data egress capability: sensitive prompts or internal data can be transmitted to a remote service, and a misconfigured or malicious API_URL could exfiltrate data or route requests to an untrusted host.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal