ClawHub
Back to skill

Security audit

browser-agent-tool-free

Security checks across malware telemetry and agentic risk

Overview

This is a coherent browser automation skill, but users should handle saved sessions, screenshots, and the global CLI dependency as sensitive.

Install only if you trust the referenced agent-browser npm package and need agent-driven browser automation. Use it on user-approved sites, protect auth state files like passwords, avoid committing screenshots or session JSON, and prefer test or least-privileged accounts for admin/user session examples.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger condition is so broad that the skill could be invoked for almost any AI/agent/LLM-related task, including contexts where browser automation is unnecessary or risky. In an agent platform with exec/write/read privileges, overbroad auto-invocation increases the chance of unintended browsing, data access, and unsafe actions on external sites.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
Documenting state save/load for cookies and storage without warning about sensitive session material encourages unsafe handling of authentication artifacts. Saved browser state can contain active sessions, tokens, and account data, so careless storage, reuse, or sharing may lead to account takeover or unauthorized access.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The examples instruct loading authenticated state files for admin and user sessions but provide no warning that these files may contain privileged credentials or active sessions. In this skill context, multi-session browser automation makes misuse especially dangerous because a leaked or reused admin-auth.json could grant direct privileged application access.

Missing User Warnings

Low
Confidence
85% confidence
Finding
Screenshot and PDF export features can capture visible secrets, personal data, internal dashboards, and other confidential page contents, yet the documentation presents them without any caution. In an automation skill intended for agent use, automatic export increases the chance that sensitive material is persisted to disk or shared in logs/artifacts unintentionally.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.