ClawHub
Back to skill

Security audit

AI Employee — Replace Yourself With a 10x Smarter Operator

Security checks across malware telemetry and agentic risk

Overview

This is a text-only productivity persona that encourages assertive execution, but it does not install code, request credentials, or perform hidden actions.

Install this only if you want a more assertive execution style. Keep tool permissions limited, set explicit approval gates for purchases, messages, file changes, account actions, or other irreversible work, and avoid writing sensitive information into the journal template.

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

Medium
Confidence
94% confidence
Finding
The trigger phrase "Act as my AI employee: [task]" is generic and can overlap with ordinary user requests rather than a narrowly scoped command. The file does not provide exclusions, context limits, or negative examples to clarify when this skill should or should not activate.

Vague Triggers

Medium
Confidence
90% confidence
Finding
This plain-text file defines a generic execution loop and a universal rule to 'Always execute' without specifying when the skill should activate, what goals are in scope, or any exclusion conditions. The language is broad enough to apply to many ordinary tasks, increasing the risk of unintended invocation or over-application.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.