ClawHub
Back to skill

Security audit

playwright-cli-cloakbrowser

Security checks across malware telemetry and agentic risk

Overview

This skill is openly designed for stealth browser automation with reused logins and anti-bot workarounds, which creates review-worthy risk even though the behavior is not hidden.

Install only if you have authorization for the accounts, sites, and data involved. Treat the persistent profiles as sensitive account stores, protect the Manager API, avoid exposing it publicly, and do not use the network-inspection or block-retry guidance to evade a service's access controls or collect data beyond the task you are allowed to perform.

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
86% confidence
Finding
The skill explicitly instructs operators to inspect authenticated API traffic and fetch same-origin endpoints from a logged-in browser context. That enables access to private account data, tokens, and backend responses tied to a real user session, yet the documentation provides no privacy, authorization, or data-handling guardrails. In a skill centered on persistent stealth profiles and session reuse, this meaningfully increases the risk of unauthorized data access or collection.

Ssd 4

High
Confidence
95% confidence
Finding
This section normalizes attaching to persistent stealth browser profiles, reusing logged-in sessions, passing anti-bot protections, and extracting data from authenticated contexts. The combination of anti-detect fingerprinting, persistent cookies, and instructions to avoid fresh browsers materially supports defense evasion and covert access patterns rather than ordinary test automation. In context, the skill is more dangerous because it is purpose-built to preserve identity state and bypass protections that sites use to distinguish normal from automated access.

Ssd 2

High
Confidence
93% confidence
Finding
The skill gives operational tactics for reverse-engineering site APIs, capturing network traffic, polling around challenge pages, and retrying with fresh exit IPs on blocking responses. Even without explicit exploit language, these are defense-evasion and covert extraction techniques that help an operator circumvent rate limits, bot detection, and access controls around logged-in data. The surrounding stealth-browser context makes the risk more acute because the advice is directly actionable against protected real-world targets.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.