ClawHub
Back to skill

Security audit

公司情报侦察

Security checks across malware telemetry and agentic risk

Overview

This skill openly performs sales-intelligence work, but it can probe email addresses and create local and CRM records without clear approval steps.

Review before installing. Use only for authorized business research, keep SMTP probing disabled unless explicitly approved for the exact targets, and require a preview before creating local files or OKKI CRM records.

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 (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger condition is extremely broad: any user who provides a company URL or name and asks for profiling, background research, contacts, or emails can activate a workflow that performs reconnaissance, contact discovery, email inference, SMTP validation, local file creation, and CRM writes. This increases the chance of unintended activation on ordinary business-research requests and can enable large-scale collection of personal contact data without clear user confirmation or scope checks.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly performs persistent local writes and creates records in an external CRM (OKKI) but does not prominently require user authorization, preview, or consent for the data being stored and transmitted. In context, this is more dangerous because the workflow also harvests and validates individual contact emails, so the writes can persist potentially sensitive or improperly collected personal/business intelligence into internal systems.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.