ClawHub

opencli-browser

Security checks across malware telemetry and agentic risk

Overview

This is a coherent browser automation skill, but it gives an agent broad control of logged-in browser sessions and can persist captured API responses without enough safety guidance.

Install only if you intentionally want an agent to control a real Chrome window. Prefer a dedicated browser profile or test account, supervise sensitive sessions, close the automation window when done, clear the opencli network cache after sensitive work, and require explicit confirmation before the agent submits forms, posts content, changes account settings, buys anything, or saves extracted data.

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
94% confidence
Finding
The skill explicitly encourages use of browser network capture to extract API responses from live, potentially logged-in sessions, but it does not warn that captured traffic may include session cookies, auth tokens, personal data, or internal API payloads. In an agent-facing skill, that omission is dangerous because it normalizes broad inspection of authenticated traffic and increases the chance of inadvertent sensitive-data exposure or over-collection.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill documents that captured network bodies are cached on disk in a persistent directory, yet provides no warning that responses from logged-in sessions may contain credentials, tokens, financial data, or other sensitive records. Persisting such data without prominent caution, retention guidance, or cleanup instructions materially increases the risk of local disclosure and unintended reuse by later processes or users.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal