ClawHub

InfoDashboard

Security checks across malware telemetry and agentic risk

Overview

InfoDashboard is a coherent setup guide, but it gives an agent access to database credentials, private-network tunneling, external LLM processing, and generated Docker dashboards with some important scoping gaps.

Install only if you trust the external InfoDashboard repository. Pin or inspect the repo before installing, use an isolated environment, use read-only least-privilege database credentials, avoid secrets or PII in prompts, bind or firewall the service appropriately, review generated code before relying on it, and stop tunnels and containers when finished.

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 (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The flow instructs the agent to immediately POST arbitrary user requests to a generation endpoint while the server implicitly reads database credentials from .env.local, but it provides no warning about sensitive data handling, outbound transmission, or least-privilege credential use. In this skill context, that is materially risky because natural-language prompts may contain proprietary business data and the generation pipeline explicitly involves multiple agents, database schema access, and LLM processing, increasing the chance of unintended disclosure or overbroad data access.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The document exposes a DELETE endpoint for removing dashboards with only a brief confirmation note and no warning about user impact, data loss of generated artifacts, or the need to verify the target dashboard. In this operational skill, destructive actions are state-changing and could disrupt active users or remove a needed dashboard if the agent or operator targets the wrong dashboard ID.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal