← Back to Skills Marketplace
12283
Downloads
22
Stars
48
Active Installs
1
Versions
Install in OpenClaw
/install ws-agent-browser
Description
浏览器智能控制。自动化操作、截图、填表、数据抓取。
Usage Guidance
Install only if you intentionally want an agent to control a browser. Give specific URLs and task boundaries, require confirmation before submitting forms or acting in logged-in accounts, avoid sensitive sessions when possible, set explicit limits for monitoring jobs, and delete saved screenshots or scraped data when no longer needed.
Capability Assessment
Purpose & Capability
The advertised browser actions are coherent with the stated purpose: opening pages, screenshots, clicking, typing, form filling, scraping, login workflows, monitoring, and script evaluation.
Instruction Scope
The skill uses broad triggers such as browser, automation, screenshot, form filling, and scraping, and it tells users they can give vague commands like opening a webpage, screenshotting, or filling a form. That is under-scoped for a skill that can click, type, scrape, authenticate, and run browser scripts.
Install Mechanism
The artifact is a single SKILL.md file with no executable scripts, binaries, package installs, or install-time commands.
Credentials
Using a local browser and storing screenshots, recordings, scraped data, and config under disclosed workspace paths is proportionate to the feature, but those outputs may contain private page or account content.
Persistence & Privilege
The skill advertises batch operations, scheduled tasks, hourly monitoring, login authentication, and browser_evaluate script execution, but only gives a generic statement that sensitive operations require confirmation without defining scope, duration, or confirmation thresholds.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ws-agent-browser - After installation, invoke the skill by name or use
/ws-agent-browser - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Agent Browser v1.0.0 初始发布
- 智能控制浏览器，实现自动化操作、截图、表单填写和数据抓取
- 支持定时任务、批量操作与页面监控
- 集成常用浏览器（Chrome/Brave/Edge/Chromium）
- 提供多项页面操作及数据导出能力
- 明确操作日志与安全规范
Metadata
Frequently Asked Questions
What is Agent Browser?
浏览器智能控制。自动化操作、截图、填表、数据抓取。 It is an AI Agent Skill for Claude Code / OpenClaw, with 12283 downloads so far.
How do I install Agent Browser?
Run "/install ws-agent-browser" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Agent Browser free?
Yes, Agent Browser is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Agent Browser support?
Agent Browser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Agent Browser?
It is built and maintained by fhbillwer (@fhbillwer); the current version is v1.0.0.
More Skills