ClawHub

Temp Agent Browser

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate browser automation skill, but it needs Review because it can act inside logged-in browser sessions and save or expose sensitive browser data without enough scoping guidance.

Install only if you trust the external agent-browser package. Use a dedicated or throwaway browser session for automation, avoid sensitive accounts unless necessary, approve high-impact submissions/uploads/account changes yourself, store any auth state or capture files in protected locations, and clear cookies/storage after use.

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 (2)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly documents `set credentials` and saved session state commands without any warning about storing, replaying, or exposing secrets such as passwords, cookies, and auth tokens. In an agent context, this can normalize unsafe handling of authentication material and lead to credential leakage through logs, shell history, persisted files, or reuse across tasks.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill describes screenshots, PDFs, video recording, trace capture, and file output features without warning that these artifacts may capture sensitive page contents, credentials, tokens, PII, or internal application data. Because the tool is intended for autonomous browser use, agents may create local files containing sensitive information that persist beyond the task and are easy to exfiltrate or mishandle.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal