
Agent Browser
- 1.4k installs
- skills.volces.com
CLI tool that executes browser automation commands: navigate, interact with DOM elements, capture snapshots and screenshots, for tasks requiring browser control.
About
Agent Browser is a Playwright-based CLI tool from Vercel for browser automation. It provides commands to navigate, interact with page elements (click, fill, select), capture snapshots of accessibility trees, and take screenshots. Developers use it for sign-in automation, form submission workflows, and information extraction tasks. Key workflows include opening URLs, getting page structure via snapshot, locating elements by ref or CSS selector or ARIA role, performing interactions, and capturing results. Typical scenarios are daily check-in tasks, multi-field form filling, and data collection from web pages.
- Playwright-based CLI with open/click/fill/select/check commands for page control
- Accessibility tree snapshot output provides ref-based element targeting (@eXX) plus CSS and ARIA role finders
- Screenshot and annotated screenshot commands for visual capture and debugging
- Chainable workflows: navigate, interact, wait, snapshot, and close within single browser session
- Cron-compatible shell scripts for scheduled automation (e.g., daily sign-in with timestamp logging)
Agent Browser by the numbers
- 1,446 all-time installs (skills.sh)
- +1 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #817 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
agent-browser capabilities & compatibility
- Capabilities
- navigate pages (open, back, forward, reload) · interact with elements (click, fill, select, che · capture page state (snapshot, screenshot, annota · extract information (get text, get html, get val · locate elements by ref, css selector, or aria pr · maintain browser session across multiple command
- Works with
- playwright
- Use cases
- web scraping · testing
- Platforms
- macOS · Windows · Linux · WSL
- Runs
- Runs locally
- Pricing
- Free
What agent-browser says it does
Vercel 出品的浏览器自动化 CLI,基于 Playwright,比标准浏览器工具更快更灵活。
npx skills add https://github.com/skills.volces.com --skill agent-browserAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1.4k |
|---|---|
| Security audit | 1 / 1 scanners passed |
| Repository | skills.volces.com ↗ |
What it does
Automate browser interactions such as login, form filling, screenshots, and web scraping via CLI commands.
Who is it for?
Automating sign-in flows, form submission, web scraping, screenshot capture, page structure inspection.
Skip if: Complex visual testing requiring pixel-perfect comparison; performance-critical browser automation; tasks not needing human-readable page snapshots.
When should I use this skill?
User requests browser automation, form filling, page navigation, element interaction, or web content extraction via CLI or agent workflows.
What you get
Reduced manual browser work; reliable, auditable automation via simple CLI commands; reusable scripts for scheduled tasks.
By the numbers
- Supports 10+ interaction commands: open, click, dblclick, fill, type, select, check, uncheck, press, snapshot
- Element location via 3 methods: ref-based (@eXX), CSS selectors, ARIA roles/labels/text/placeholder
- Screenshot variants: standard and annotated (with element markers)
Files
Agent Browser
Vercel 出品的浏览器自动化 CLI,基于 Playwright,比标准浏览器工具更快更灵活。
快速开始
agent-browser open <url> # 打开网页
agent-browser snapshot # 获取页面可访问性树
agent-browser click @<ref> # 点击元素(用ref引用)
agent-browser fill @<ref> "内容" # 填入内容
agent-browser close # 关闭浏览器常用命令
导航
agent-browser open <url> # 打开URL(别名:goto, navigate)
agent-browser back # 后退
agent-browser forward # 前进
agent-browser reload # 刷新交互
agent-browser click <sel> # 点击
agent-browser dblclick <sel> # 双击
agent-browser fill <sel> "text" # 填入(清空后填)
agent-browser type <sel> "text" # 输入(追加)
agent-browser select <sel> <value> # 选择下拉选项
agent-browser check <sel> # 勾选复选框
agent-browser uncheck <sel> # 取消勾选
agent-browser press <key> # 按键(Enter, Tab, Escape等)获取信息
agent-browser snapshot # 获取可访问性树(推荐)
agent-browser get text <sel> # 获取文本
agent-browser get html <sel> # 获取HTML
agent-browser get value <sel> # 获取输入值
agent-browser get title # 获取页面标题
agent-browser get url # 获取当前URL
agent-browser screenshot [path] # 截图
agent-browser screenshot --annotate # 带标注的截图元素定位
通过 snapshot 输出的 ref(如 @e14)直接引用:
agent-browser click @e14
agent-browser fill @e13 "hello"或使用 CSS 选择器:
agent-browser click "#submit"
agent-browser fill "input[name='email']" "test@test.com"或使用 ARIA 角色查找:
agent-browser find role button click --name "Submit"
agent-browser find text "Sign In" click
agent-browser find label "Email" fill "test@test.com"
agent-browser find placeholder "Search" type "query"典型工作流
1. 签到任务
# 打开登录页
agent-browser open <签到页面URL>
# 获取页面结构
agent-browser snapshot
# 点击登录/签到按钮(用实际ref替换 @eXX)
agent-browser click @eXX
# 等待页面加载
sleep 2
agent-browser snapshot2. 填表任务
agent-browser open <表单URL>
agent-browser snapshot
# 填入各字段
agent-browser find label "用户名" fill "myuser"
agent-browser find label "密码" fill "mypassword"
agent-browser find role button click --name "提交"3. 定时签到(配合cron)
创建脚本 ~/.openclaw/scripts/daily-checkin.sh:
#!/bin/bash
agent-browser open <签到URL>
sleep 2
agent-browser find role button click --name "签到"
agent-browser screenshot /tmp/checkin_$(date +%Y%m%d).png
agent-browser close注意事项
1. 先 snapshot 再操作 - 每次页面变化后重新获取 ref 2. 添加等待 - 页面加载需要时间,用 sleep 2 或等待 3. 保持浏览器开启 - 多个操作可以在同一浏览器会话中完成 4. 完成后关闭 - 用 agent-browser close 释放资源
依赖安装
如果 agent-browser 未安装:
npm install -g agent-browser
agent-browser installRelated skills
FAQ
How do I locate elements on a page?
Use snapshot to get accessibility tree with refs (@eXX), or use CSS selectors or ARIA role finders (find role button, find label, find text, find placeholder).
Can I run automation tasks on a schedule?
Yes, create a shell script and schedule via cron. Example: ~/.openclaw/scripts/daily-checkin.sh with agent-browser commands.
Do I need to manage the browser session manually?
Minimal: use open to start, snapshot to inspect, run interactions, then close. Multiple commands share one session until close is called.
Is Agent Browser safe to install?
skills.sh reports 1 of 1 security scanners passed. Review the Security Audits panel on this page before installing in production.