Agent Browser CLI

使用 agent-browser CLI 进行浏览器自动化。用于签到、填表、截图、信息抓取等需要控制浏览器的任务。触发条件：(1) 用户要求自动化浏览器操作 (2) 需要签到、填表、点击按钮 (3) 需要抓取网页内容作为研究素材

Security Scan

The SKILL.md and requirements are internally consistent for a CLI-based browser automation helper, but the package has unknown provenance and the recommended global npm install raises moderate installation risk — inspect before installing.

Purpose & Capability

The name/description (browser automation, sign-ins, form-filling, scraping) matches the instructions and commands in SKILL.md; there are no unrelated env vars, binaries, or config paths requested.

Instruction Scope

Instructions stay within browser automation: opening pages, clicking, filling, snapshots, screenshots and creating a small cronable script. They do recommend writing a script under ~/.openclaw/scripts and using screenshots and page scraping (which is expected for this purpose), so users should be aware these actions can capture and store sensitive page content if misused.

Install Mechanism

No install spec in the skill package itself, but SKILL.md recommends `npm install -g agent-browser` and `agent-browser install`. Installing a global npm package can run arbitrary install/postinstall scripts and persists code on disk — a moderate-risk, common distribution method. No direct download URLs are provided.

Credentials

The skill declares no required environment variables, credentials, or config paths. The example uses plaintext fills (e.g., password) but does not request secrets from the environment.

Persistence & Privilege

always is false and the skill is user-invocable; autonomous invocation is enabled by default (normal). The skill does suggest creating a user script and possibly a cron job, which is standard for scheduled automation but is user-controlled rather than requiring system privileges.

Assessment

This skill appears coherent for browser automation, but take these precautions before installing: (1) The package has no homepage/source listed — inspect the npm package before running a global install (use `npm view agent-browser`, `npm pack`, or review its repository) because npm packages can run arbitrary code at install time. (2) Prefer installing in a contained environment (non-root user, container, or dedicated VM) if you want to reduce risk. (3) Review any cron scripts you create (e.g., ~/.openclaw/scripts/*) and avoid storing secrets in plaintext inside them. (4) Be mindful that screenshots and page scraping can capture sensitive data; restrict automated workflows and require explicit confirmation for actions that sign in or submit forms. (5) If you need higher assurance, ask the publisher for source code or a verified homepage and re-run the evaluation once provenance is known.

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']" "[e-mail]"

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 "[e-mail]"
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 snapshot

2. 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 install