ClawHub
Back to skill

Security audit

Browser Healthcheck

Security checks across malware telemetry and agentic risk

Overview

This browser troubleshooting skill is mostly coherent, but its repair path can forcibly stop local processes without enough safeguards.

Review before installing if you expect the assistant to run repair steps automatically. Use diagnostic mode first, and only allow --fix or taskkill after confirming the PID belongs to the intended OpenClaw browser process, not your active Chrome or another local service.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
if fix:
                print(f"Attempting to kill process {pid}...")
                try:
                    subprocess.run(['taskkill', '/F', '/PID', str(pid)], 
                                 check=True, capture_output=True)
                    print_ok(f"Process {pid} killed")
                    time.sleep(2)
Confidence
88% confidence
Finding
subprocess.run(['taskkill', '/F', '/PID', str(pid)], check=True, capture_output=True)

Intent-Code Divergence

Medium
Confidence
84% confidence
Finding
The script presents itself as a diagnostic health check, but it also contains destructive remediation logic that can forcibly kill processes. That mismatch increases the chance that users or agents invoke it assuming read-only behavior, making the destructive path more dangerous in practice.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill includes forceful process-termination guidance (`taskkill /F /PID <pid>`) and restart instructions without any warning to verify the target process, save browser state, or consider collateral impact. In a troubleshooting skill, this can cause unintended termination of the user's own Chrome session or other processes bound to the port, leading to data loss, disrupted work, or unsafe operational behavior.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The `--fix` path performs an immediate forced kill once a PID is identified, without interactive confirmation or stronger validation that the process is safe to stop. This creates avoidable risk of accidental service disruption, especially when run non-interactively by an agent or script.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.