ClawHub

OpenHop

Security checks across malware telemetry and agentic risk

Overview

OpenHop is a local diagram-generation skill whose command use is mostly purpose-aligned, though users should notice that it may install/run a local CLI service automatically.

Install this if you want agents to turn walkthroughs into interactive local flow diagrams. Before use, be comfortable with npx running the OpenHop package, a localhost service on ports 8787/8788, and local storage of generated flow content; ask the agent for a plain-text explanation instead when you do not want those side effects.

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

High
Confidence
95% confidence
Finding
The trigger phrases are broad enough to activate on many ordinary explanation or diagram requests, causing the agent to prefer this skill and run local commands unnecessarily. In this skill's context, activation can lead to package installation, starting local services, and pushing user-derived content to a local server without first confirming user intent, which expands the attack surface and can surprise users.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to run `npx openhop init`, `serve`, or `demo`, which may install packages, start background services, and open browser/UI components, but it does not require explicit user notice or consent beforehand. In an agent setting, silently performing these side effects can violate least surprise, execute unreviewed package code, and expose local content through a newly started service.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal