ClawHub
Back to skill

Security audit

broswer use skill

Security checks across malware telemetry and agentic risk

Overview

This browser automation skill is openly described, but it gives agents broad control over Chrome, page contents, uploads, downloads, and tool updates without enough user-control guidance.

Install only if you are comfortable giving an agent command-line control over a connected Chrome browser. Avoid using it on banking, work, admin, or other sensitive logged-in pages unless you explicitly approve each action, and require confirmation before uploads, form submissions, purchases, deletes, screenshots, LLM analysis, downloads, skill script execution, or self-updates. Static scan was clean and VirusTotal was pending, so this Review verdict is based on the artifact’s broad, under-scoped authority rather than malware evidence.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill explicitly supports screenshots, downloads, file uploads, and shell-driven browser automation, but it provides no safety guidance about sensitive data exposure, local file writes, or user-consent boundaries. In a browser-control skill, these capabilities can directly access authenticated sessions, page contents, and local files, so the absence of warnings or guardrails materially increases the risk of unintended data exfiltration or destructive actions.

Self-Modification

High
Category
Rogue Agent
Content
```bash
browser-cli status    # Check daemon + extension connection
browser-cli update    # Self-update to latest version
```

---
Confidence
93% confidence
Finding
Self-update

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.