ClawHub

Sales AI Agent Starter

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a prompt-only guide for building business AI assistants, with expected but sensitive deployment and data-handling topics that need careful user oversight.

Install only if you want guidance for business AI assistant implementation. Before using it with real systems, test in a sandbox, use least-privilege credentials, confirm every CRM/API write, avoid unnecessary customer-data retention, and add your own privacy/compliance and rollback procedures.

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
90% confidence
Finding
The trigger phrases are broad enough to match many ordinary requests about AI assistants, chatbots, or automation, which can cause this skill to activate outside its intended scope. Overbroad routing increases the chance of inappropriate tool use or the agent steering users toward deployment/integration workflows when they only wanted generic advice.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill encourages deployment, API integration, CRM synchronization, and data capture/storage, but it does not warn about handling sensitive business data, production-side effects, access scopes, or privacy/compliance obligations. In this context, omission of safety guardrails is risky because the skill is explicitly positioned to build and deploy operational agents that may touch customer records and live systems.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal