ClawHub

Clawhub Skill Scanner

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local scanner for reviewing OpenClaw skills, but its results should be treated as advisory rather than automatic permission to install.

Reasonable to install if you want a local heuristic scanner, but do not rely on a SAFE result as a guarantee. Review findings yourself before installing external or high-privilege skills, and be aware that the scanner can miss patterns because it uses simple line-based regex checks.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The whitelist runs before any suspicious-pattern checks and skips entire lines containing comments, docstring markers, API URLs, or localhost text. That creates a straightforward evasion path where malicious code or dangerous strings can be hidden on otherwise whitelisted lines, causing the scanner to miss them and undermining its core security purpose.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrases are broad enough to activate on routine discussion of installing or adding skills, which can cause the scanner to run unexpectedly or intercept unrelated workflows. While not directly malicious, broad activation can create denial-of-service style friction, prompt hijacking, or workflow confusion in agent environments where skills auto-trigger.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| Score | Level | Action |
|-------|-------|--------|
| 0-20 | 🟢 SAFE | Auto-approve |
| 21-50 | 🟡 CAUTION | Review findings |
| 51-80 | 🔶 DANGER | Detailed review required |
| 81-100 | 🔴 BLOCKED | Do NOT install |
Confidence
69% confidence
Finding
Auto-approve

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal