ClawHub

Data Toolkit

Security checks across malware telemetry and agentic risk

Overview

This skill is a local data toolkit whose file changes match its stated purpose, but users should avoid accidental in-place data loss when cleaning files.

Install only if you are comfortable running local Python scripts on your datasets. Use --output for cleaning commands, keep backups of important files, and test cleaning rules on a small sample before running them on original data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill advertises and demonstrates file read/write behavior through conversion and cleaning commands, but the metadata declares no corresponding permissions or safety constraints. This creates a trust and authorization gap: an agent may process or overwrite local files without users being clearly informed of the capability scope.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The cleaning features include duplicate removal, null removal/replacement, normalization, and a full cleanup pipeline, all of which can irreversibly alter datasets. Because the documentation does not prominently warn about destructive effects or encourage backup/output-file usage, users or agents may run these operations in ways that silently lose or corrupt important data.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The tool writes back to the input path by default when --output is not provided, which can silently destroy the original dataset if cleaning rules behave unexpectedly or are misconfigured. In an agent/workflow context, this is more dangerous because file operations may be triggered automatically on user-supplied paths, turning a usability issue into unintended data loss or corruption.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal