ClawHub
Back to skill

Security audit

Skill Install Bridge

Security checks across malware telemetry and agentic risk

Overview

This is a local helper that generates install-copy files for published skills, with no evidence of hidden network access, credential use, persistence, or destructive behavior.

Install this if you want a helper that creates local install snippets for published skills. Review the generated Markdown, HTML, JSON, and social text before publishing, provide real skill metadata, and use --force only when you intentionally want to overwrite existing generated files.

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 (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill metadata declares no permissions, yet the workflow explicitly invokes a bundled Python script that writes output files and may access network resources via referenced public URLs or provenance data. This mismatch can mislead reviewers and users about the skill's real capabilities, reducing informed consent and weakening sandboxing or policy enforcement around file-write and network access.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.