ClawHub
Back to skill

Security audit

Agent Browser Clawdbot Local

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate Playwright browser automation skill, but it can handle browser login state and cookies that should be treated as sensitive.

Use this skill mainly with test accounts or sessions you intentionally create for automation. Treat saved state files such as auth.json, cookies, and localStorage values like credentials: do not share or commit them, delete them after use, and avoid attaching to a personal logged-in browser unless you explicitly want the agent to operate in that session.

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
91% confidence
Finding
The skill explicitly instructs users to save and load browser auth state, which commonly includes cookies, local/session storage, and other session artifacts that can grant account access. In an agent context, presenting this workflow without any warning about sensitivity, storage hygiene, or scope increases the risk of credential leakage, unintended reuse across tasks, and privilege crossover between sessions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill documents commands for reading and modifying cookies and browser storage, but gives no privacy or security guidance despite these interfaces often exposing authentication tokens, PII, CSRF material, and application secrets. Because this skill is designed for AI agents, omission of handling constraints makes accidental exfiltration, unsafe logging, and unauthorized state manipulation more likely.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.