ClawHub
Back to skill

Security audit

Anti-Detect Browser

Security checks across malware telemetry and agentic risk

Overview

The skill is a clearly disclosed but dual-use anti-detect browser integration, with meaningful guidance to limit it to authorized QA, scraping, and account-management scenarios.

Install only for authorized browser automation, QA, ad verification, public-data collection, or accounts you own or administer. Pin package versions, understand that a closed-source browser kernel will be downloaded, keep API keys and proxy URLs in environment-managed secrets, and protect the persistent profile directory because it can contain live sessions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (1)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The skill advertises a very broad set of trigger phrases, including generic and abuse-adjacent terms such as multi-accounting, account association evasion, residential proxy, and Playwright stealth. This can cause the skill to be invoked in contexts beyond legitimate QA or scraping, including requests aimed at evading bot detection or operating multiple accounts, increasing the chance an agent selects a high-risk capability for ambiguous user prompts.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.