Back to skill

Security audit

Browser VM Setup

Security checks for vulnerabilities and agentic risk

Overview

This skill is a setup guide for running a browser agent on a VM, with sensitive behavior like sessions, screenshots, shell-capable dashboard access, and retention disclosed and scoped to user-controlled infrastructure.

Before installing, be comfortable that this guide is for a VM you control and that it involves privileged setup, browser session persistence, screenshots/logs that may contain personal data, and a local dashboard with shell-level power. Keep the dashboard loopback-only, use SSH tunneling, avoid public exposure, install and verify retention cleanup, and do not use proxy or residential egress paths unless you have explicit authorization for the target.

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.