ClawHub
Back to skill

Security audit

Agent Ui

Security checks across malware telemetry and agentic risk

Overview

This is a coherent React/Next.js agent UI integration guide with normal third-party setup risks, not evidence of hidden or malicious behavior.

Before installing, review the remote component and @inferencesh/sdk package, pin versions where practical, keep INFERENCE_API_KEY server-side only, protect the proxy route with authentication and rate limits, avoid logging secrets or sensitive prompts/files, and restrict browser-side tools to actions users explicitly approve.

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 list includes broad, generic phrases such as 'ai assistant ui', 'copilot ui', 'react agent', and 'chat agent' that can match many unrelated requests. In an agent/skill-routing system, this can cause the skill to activate unexpectedly, steering users toward a component that enables proxying, tool execution, and agentic behaviors in contexts where it was not specifically requested.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documentation instructs users to create a proxy route and set an API key, but it does not warn about sensitive credential handling, server-only storage, request scoping, logging risks, or what data may be transmitted through the proxy. Because this skill is specifically for agentic UI features with streaming, tools, and file/image handling, the absence of those warnings increases the chance that integrators will expose secrets or forward sensitive user data without appropriate safeguards.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.