ClawHub

OnboardAI-新员工入职引导系统

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local onboarding dashboard, but it should not be used with real HR data or real access secrets without tightening privacy and credential handling.

Install only if you are comfortable with a local browser-based HR onboarding tool. Do not enter real employee records unless localStorage and local JSON exports are acceptable for your organization, and replace the built-in WiFi/password examples with secure, authenticated onboarding instructions such as SSO or one-time activation links.

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 (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger word at this location is a generic HR-related term that can match many ordinary requests beyond the skill's intended dashboard workflow. Overly broad activation can cause the agent to invoke this skill in unrelated contexts, leading to inappropriate file access, unwanted UI actions, or accidental exposure of locally stored onboarding data.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger list contains many broad phrases such as general onboarding and HR system terms, which increases the chance of unintended activation for routine HR conversations. In this skill’s context, unintended activation is more concerning because the skill is allowed to use Bash, Read, and Write tools and is designed to open a local interactive asset, so misrouting could lead to unnecessary tool use or data handling.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The export function writes a JSON report containing employee personal data, onboarding status, and satisfaction records directly to a downloadable file without any warning, minimization, or access control. In an HR onboarding context, this increases the risk of accidental disclosure of personnel data to unauthorized recipients or insecure storage locations.

Missing User Warnings

High
Confidence
97% confidence
Finding
The embedded knowledge base includes direct disclosure of sensitive access information, including Wi‑Fi credentials and account/password handling details, and the chat surfaces it to any user of the page. In an onboarding skill, this materially increases the chance of credential leakage, shoulder-surfing, unauthorized reuse, or redistribution of internal access details.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal