ClawHub
Back to skill

Security audit

Browser CDP Tailnet

Security checks across malware telemetry and agentic risk

Overview

This skill is not malicious, but it gives agents broad control of a shared remote browser without enough privacy, isolation, or user-control guidance.

Install only if you trust the tailnet endpoint and everyone who can access the shared browser. Use a dedicated non-sensitive browser profile, avoid logging into important accounts, and require explicit user approval before submitting forms, changing account data, or using authenticated sessions.

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 (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly exposes and instructs use of a shared remote Chromium instance over a tailnet-accessible CDP endpoint, but provides no safeguards around shared-session privacy, cross-user data exposure, or trust boundaries. A browser connected through CDP typically grants powerful control over tabs, cookies, storage, navigation, and page content, so directing agents to use a shared instance can leak credentials, browsing data, or active sessions between users or agents.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.