ClawHub
Back to skill

Security audit

CEO / Chief Executive Officer

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable CEO advisory skill with scoped local memory for company context and clear human sign-off boundaries for consequential actions.

Before installing, be aware that this skill may keep CEO-related preferences and company facts such as cash, burn, board members, customers, investors, and red lines in local files under ~/Clawic/data/ceo/. Review or delete those files if you do not want that context reused later; major legal, financial, personnel, public, board, M&A, layoff, or shutdown actions should still receive human and professional 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

Medium
Confidence
91% confidence
Finding
The catch-all trigger "Anything else" makes the skill eligible for invocation outside its intended company-level CEO scope, which can cause misrouting and over-broad activation. In this skill, that risk is amplified because it presents authoritative strategic, financial, personnel, and crisis guidance; accidental use in narrower functional or sensitive contexts could bypass more appropriate specialized skills or human review expectations.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly directs the agent to read and persist sensitive company and user context across `config.yaml`, `company.md`, and `memory.md` without any user-facing disclosure or consent step. Because this CEO skill handles high-sensitivity business data such as cash, burn, board roster, investors, customers, and strategic red lines, silent persistence increases privacy and confidentiality risk and could expose data to later prompts or other skills.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.