ClawHub
Back to skill

Security audit

Hybrid DB Health

Security checks across malware telemetry and agentic risk

Overview

This health-check skill is mostly purpose-aligned, but it can expose local API keys and run live third-party database/API checks without clear opt-in.

Review before installing. Do not run the manual grep command as written unless you are prepared for API keys to appear in logs or chat output. Prefer editing the check to verify only whether keys are present, redact any error output, and require confirmation before running query.py against OpenAI/Pinecone.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
This line performs a live API-backed query using locally stored OpenAI and Pinecone credentials, which can cause outbound network access and billable requests without clearly warning the user. In an agent-skill context, silent external calls are more dangerous because they may disclose metadata, trigger unintended third-party processing, or incur cost during what appears to be a simple health check.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.