ClawHub
Back to skill

Security audit

find-skills-wzr-999

Security checks across malware telemetry and agentic risk

Overview

This skill is for finding OpenClaw skills, but it also encourages broad triggering and global auto-confirmed installation of third-party skills without clear user approval.

Review before installing. Use this only when you intentionally want to search for or install OpenClaw skills, and do not run the global `-g -y` install command unless you have checked the source, understand it will persistently modify your skill environment, and explicitly approve that installation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The metadata description says the skill should be used for broad prompts like 'how do I do X' or whenever a user expresses interest in extending capabilities. Those triggers are so general that this skill could activate in many unrelated conversations and steer users toward searching/installing third-party skills when they did not explicitly ask for that, increasing the chance of unnecessary package installation and supply-chain exposure.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The 'When to Use This Skill' section lists ambiguous conditions such as any specialized capability request, interest in extending capabilities, or wishing for help in a domain. In practice, this can cause over-triggering and push users into a workflow that searches for and installs external skills, which broadens attack surface and may bypass the safer default of answering directly.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to offer installation using 'npx skills add <owner/repo@skill> -g -y', which performs a global install and suppresses confirmation prompts, but it does not prominently warn about either behavior or require explicit user approval. Because skills may come from GitHub or other external sources, this materially increases the risk of silent installation of untrusted code and persistent modification of the user's environment.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.