ClawHub
Back to skill

Security audit

Skill Publisher

Security checks across malware telemetry and agentic risk

Overview

This skill is a public-publishing workflow that is mostly disclosed, but it grants broad release-oriented authority with weak confirmation and a hardcoded publisher namespace.

Install only if you intentionally want this publisher-specific workflow. Before running it, confirm the exact skill folder, use --dry-run first, verify the cleaned bundle contains no secrets, and be aware that the generated install commands and namespace are hardcoded to mebusw rather than your own account.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The activation criteria are broad enough to trigger on generic requests to 'publish' or 'ship' a skill folder to public destinations, which can cause the agent to begin a release-oriented workflow without sufficient scoping or confirmation. In a skill that performs file preparation, documentation generation, and potential handoff toward public distribution, overbroad triggering increases the chance of unintended disclosure or accidental publication preparation.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The usage guidance includes vague examples and a fallback condition that effectively says to use this skill whenever a skill folder goes from local to public. That broad scope makes accidental invocation more likely in sensitive contexts, especially because later steps create files, duplicate directories, and prepare artifacts for external release.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.