ClawHub

hr-admin-assistant

Security checks across malware telemetry and agentic risk

Overview

This HR/admin skill is mostly local and purpose-aligned, but it asks the agent to persistently broaden its own activation triggers, so users should review it before installing.

Install only if you are comfortable with a broad HR/admin drafting assistant handling sensitive employee, policy, legal-support, and finance-support material. Before use, narrow or remove the generic triggers, remove the instruction that appends new trigger words to SKILL.md unless changes require explicit approval, and treat legal or HR compliance outputs as drafts for qualified review.

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

High
Confidence
95% confidence
Finding
The trigger list contains very broad phrases such as '帮我写', '写一个', and '做一个方案', which are common in many unrelated conversations. This can cause the skill to activate outside HR/admin contexts, increasing the chance that it reads internal references or steers responses inappropriately when another skill or safer default behavior should apply.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation criteria describe a very wide class of 'back-office' requests, including drafting, analysis, process design, policy creation, and document organization, without firm boundaries. Such ambiguity can cause over-selection of this skill for sensitive or unrelated tasks, leading to policy overreach, disclosure of internal templates, or incorrect handling of specialized legal/financial matters.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal