Security audit

浏览器CLI工具-免费版

Security checks for vulnerabilities and agentic risk

Overview

This is a browser automation helper, but it should be reviewed because it encourages password form submission and recurring unattended sign-ins without enough guardrails.

Install only if you are comfortable giving an agent browser-control and command-execution ability. Avoid inline passwords, confirm before submitting forms or account actions, do not schedule unattended sign-ins unless you understand the site's rules, and clean up screenshots that may contain private information.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

High

Confidence: 88% confidence
Finding: Overly broad trigger conditions can cause an agent to invoke a browser automation skill in contexts outside its intended scope, increasing the chance of unsafe actions, unintended web interactions, or misuse on sensitive workflows. In an agent ecosystem, poor routing guidance is dangerous because it can expand the operational surface area of a tool that has execution and browser-control capability.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The examples normalize entering usernames and passwords and submitting forms without any warning, consent checkpoint, or guidance for secrets handling. In a browser-automation skill, this is more dangerous because it encourages autonomous handling of credentials and account-affecting actions, which can lead to secret exposure, unintended submissions, or actions taken on behalf of the user without sufficient confirmation.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The scheduled sign-in example promotes recurring automated account actions and local screenshot creation without warning about persistent side effects, account policy violations, or sensitive data being written to disk. In this skill's context, automated browser control plus cron scheduling materially increases risk because the behavior can continue unattended and generate files containing session, account, or page-sensitive information.

Static analysis

No suspicious patterns detected.