Security audit

skill test

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent browser automation skill, but it exposes powerful browser-session features that users should handle carefully.

Install only if you need full browser automation and trust the upstream agent-browser package. Prefer test accounts or dedicated browser sessions, avoid printing cookies or storage values, keep saved state files such as auth.json out of repositories, and delete screenshots, PDFs, recordings, traces, and saved sessions when they are no longer needed.

SkillSpector

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: 90% confidence
Finding: The skill explicitly exposes commands for reading and persisting cookies, localStorage, HTTP credentials, and session state, but provides no safety guidance about handling secrets or authenticated browser state. In an agent setting, this can lead to unintended disclosure of tokens, cookies, or reusable login state to logs, files, or downstream tools.

Missing User Warnings

Low

Confidence: 83% confidence
Finding: The documented screenshot, PDF, and recording commands create files on disk, but the skill does not warn that these artifacts may contain sensitive page contents, credentials, or internal data. In automated environments, such artifacts can be retained, uploaded, or accessed later without the user's awareness.

Static analysis

No suspicious patterns detected.