ClawHub
Back to skill

Security audit

Create-Skill

Security checks across malware telemetry and agentic risk

Overview

This is a small instruction-only skill for creating or improving OpenClaw skills, with disclosed and scoped file generation.

Install this if you want assistance drafting or revising skills. Before relying on any generated skill, review the SKILL.md content and pay extra attention to any optional .js, .sh, or package.json files the assistant proposes.

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 (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The listed trigger phrases include broad natural-language expressions such as “创建技能” and “新建技能”, which can plausibly appear in ordinary conversation and cause the skill to activate unintentionally. In an agent framework, overly broad activation can route user requests into file-generation behavior unexpectedly, increasing the chance of unintended skill creation or modification under the workspace.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.