Security audit

Agent Browser Clawdbot.Bak 2026 01 28T18:01:09+10:30

Security checks across malware telemetry and agentic risk

Overview

This skill is a powerful but straightforward browser automation guide, with sensitive browser-state features disclosed and aligned with its purpose.

Install only if you need broad browser automation. Use isolated or test sessions when possible, keep saved auth-state files private, avoid exposing cookies or localStorage in logs or chats, and only automate accounts and sites where you have permission.

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 recommends saving and loading browser auth state, which commonly includes cookies and web storage artifacts that can preserve authenticated sessions. Without any warning about sensitivity, storage handling, encryption, or scope restrictions, an agent or user may persist reusable credentials to disk and later expose them through logs, shared workspaces, or source control.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill documents commands to read and modify cookies and local storage, both of which can contain session tokens, CSRF tokens, user identifiers, and other sensitive application state. Presenting these capabilities without warning or usage constraints increases the chance that an agent will exfiltrate, overwrite, or misuse sensitive browser data during automation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.