ClawHub
Back to skill

Security audit

splitsym

Security checks across malware telemetry and agentic risk

Overview

This is a small local comment-extraction utility with disclosed configurable matching and no evidence of hidden network, credential, persistence, or destructive behavior.

Install this if you need a local helper for extracting comments from code. Be aware that custom symbol configs are regex-based, so only run it on files you are comfortable reading into the tool, and verify the package source if using the README's pip install path instead of the provided artifact.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The skill is presented as a comment/docstring extractor, but the documented support for external JSON configuration and generic symbol extraction means it can be repurposed to extract arbitrary regex-matched content from files. That mismatch is dangerous because users or orchestrators may grant access under the assumption of narrow comment-only behavior, when the actual behavior enables broader file content extraction and potential data exposure.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.