Security audit

Browser Automation

Security checks across malware telemetry and agentic risk

Overview

This browser automation skill is not clearly malicious, but it deserves review because it can run broad browser actions while retaining sessions, saving files/screenshots, and automatically using a remote browser service when keys are present.

Install only if you trust or can inspect the actual CLI package that npm install and npm link will expose. Use a dedicated browser profile and test accounts, avoid sensitive sites unless necessary, explicitly choose local versus remote mode, clear stored profile data after use, and manually confirm any form submission, account change, download, or authenticated workflow.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The example explicitly demonstrates typing a password into a browser session and notes that Chrome's user profile may preserve session cookies between runs, but it does not clearly warn about credential exposure, session persistence, or cross-task privacy leakage. In a browser automation skill, normalized examples strongly influence usage, so this can lead operators to handle real credentials unsafely or reuse authenticated sessions unintentionally.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The download example states that files are automatically written to a local directory, but it does not warn about local filesystem side effects, storage of potentially sensitive or untrusted content, or the need to validate downloaded files. In an automation context, this may cause users to trigger unintended downloads or leave unsafe artifacts on disk.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The reference documents that downloads are automatically saved to ./agent/downloads with no dialog and no file type restrictions, which creates an implicit file-write capability from untrusted web content. In a browser automation skill, this is dangerous because navigation to attacker-controlled pages can trigger drive-by downloads or deposit misleading/executable files on disk without an explicit user confirmation step.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill description is broad enough to activate on many ordinary browsing requests, which can cause this skill to be selected in situations involving sensitive sites, authenticated sessions, or private data. In combination with automation capabilities, overbroad routing increases the chance of unintended actions or data exposure without clear user awareness.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill silently switches to a remote Browserbase environment when credentials are present, without warning the user that browsing activity, page contents, and possibly session data may be sent to a third-party service. This creates a meaningful data handling and privacy risk, especially when users expect a local browser but the skill transparently uses an external provider.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.