Back to skill

Security audit

Chrome Remote Browser Guide

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only browser automation guide, but it gives an agent broad control over logged-in Chrome sessions without enough safety boundaries.

Install only if you intentionally want an agent to control Chrome through remote debugging. Prefer a separate low-privilege browser profile, close unrelated tabs, avoid sensitive pages, and require explicit confirmation before submitting forms, changing settings, posting, purchasing, deleting data, saving or sharing screenshots, or running JavaScript on authenticated sites.

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 (3)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
This skill explicitly teaches agents to interact with authenticated browser sessions and login-gated content, which can expose emails, dashboards, tokens, account data, or permit state-changing actions on behalf of the user. In context, the omission of explicit consent, scope limitation, and sensitive-data handling guidance makes the capability meaningfully dangerous rather than merely theoretical.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The guide encourages capturing screenshots of arbitrary pages, optionally saving them to disk, and returning a CDN URL, but does not warn that screenshots may contain credentials, personal data, financial information, or internal dashboards. Exporting such captures to files or remote URLs increases the risk of unintended disclosure beyond the local browser session.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
Allowing arbitrary JavaScript execution in the page context enables inspection of DOM data, local state, and potentially sensitive content from authenticated applications, and can also mutate page state or trigger actions. In this skill's context—remote control of logged-in sessions—this materially expands the ability to exfiltrate or manipulate private user data if misused.

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.