S
SkillNav

浏览器命令行

browser-cli

by buttercannfly

Control Chrome browsers from the terminal via the AIPex extension. Use this skill when the agent needs to manage browser tabs, search page elements, click buttons, fill forms, capture screenshots, or download content — all through shell commands without an MCP client.

4.5k搜索与获取未扫描2026年4月20日

安装

claude skill add --url https://github.com/openclaw/skills

必需命令行工具

nodenpm

文档

browser-cli — Terminal Browser Control

browser-cli is a command-line tool that controls Chrome browsers through the AIPex extension's WebSocket daemon. It translates shell commands into browser actions — managing tabs, clicking elements, filling forms, capturing screenshots, and more.

Architecture:

code
browser-cli ──WebSocket──▶ aipex-daemon ──WebSocket──▶ AIPex Chrome Extension ──▶ Browser APIs

The daemon auto-spawns on first use and self-terminates when idle. No manual setup beyond initial extension connection.

When to Use This Skill

Use this skill when the user wants to:

  • Control a Chrome browser from the terminal without an MCP client
  • Open, close, switch, or organize browser tabs via CLI
  • Search for page elements and interact with them (click, fill, hover)
  • Capture screenshots of browser tabs
  • Automate browser workflows in shell scripts or CI pipelines
  • Download page content as markdown or images
  • Request human input during automated browser tasks
  • Manage AIPex skills from the command line

Trigger phrases: "browser-cli", "control browser from terminal", "browser automation CLI", "click element from shell", "terminal browser control", "command line browser", "shell browser automation"

Prerequisites

  • Node.js >= 18 installed
  • AIPex Chrome extension installed and connected to the daemon
  • browser-cli installed globally: npm install -g browser-cli

First-time Setup

After installing, connect the AIPex extension to the daemon:

  1. Open Chrome → AIPex extension icon → Options
  2. Set WebSocket URL to ws://localhost:9223/extension
  3. Click Connect
  4. Verify: browser-cli status

Command Groups

tab — Manage browser tabs

bash
browser-cli tab list                         # List all open tabs
browser-cli tab current                      # Get the active tab
browser-cli tab new https://example.com      # Open a new tab
browser-cli tab switch 42                    # Switch to tab by ID
browser-cli tab close 42                     # Close a tab
browser-cli tab info 42                      # Get tab details
browser-cli tab organize                     # AI-powered tab grouping
browser-cli tab ungroup                      # Remove all tab groups

page — Inspect and interact with page content

bash
browser-cli page search "button*" --tab 123              # Search elements by glob pattern
browser-cli page search "{input,textarea}*" --tab 123    # Search multiple element types
browser-cli page screenshot                               # Screenshot active tab
browser-cli page screenshot-tab 123 --send-to-llm true   # Screenshot with LLM analysis
browser-cli page metadata --tab 123                       # Get page metadata
browser-cli page scroll-to "#main-content"                # Scroll to element
browser-cli page highlight "button.submit"                # Highlight element
browser-cli page highlight-text "p" "important"           # Highlight text in content

interact — Click, fill, hover, and type

bash
browser-cli interact click btn-42 --tab 123                         # Click by UID
browser-cli interact fill input-5 "hello world" --tab 123           # Fill input by UID
browser-cli interact hover menu-3 --tab 123                         # Hover by UID
browser-cli interact form --tab 123 --elements '[{"uid":"in-1","value":"foo"}]'  # Batch fill
browser-cli interact editor editor-1 --tab 123                     # Get editor content
browser-cli interact upload --tab 123 --file-path /path/to/file    # Upload file
browser-cli interact computer --action left_click --coordinate "[500,300]"  # Pixel-level click

download — Save content locally

bash
browser-cli download markdown --text "# Notes" --filename notes    # Save as markdown
browser-cli download image --data "data:image/png;base64,..."      # Save image
browser-cli download chat-images --messages '[...]' --folder imgs   # Batch save images

intervention — Request human input

bash
browser-cli intervention list                                       # List intervention types
browser-cli intervention info voice-input                           # Get type details
browser-cli intervention request voice-input --reason "Need input"  # Request intervention
browser-cli intervention cancel                                     # Cancel active request

skill — Manage AIPex skills

bash
browser-cli skill list                                    # List all skills
browser-cli skill load my-skill                           # Load skill content
browser-cli skill info my-skill                           # Skill details
browser-cli skill run my-skill scripts/init.js            # Execute skill script
browser-cli skill ref my-skill references/guide.md        # Read skill reference
browser-cli skill asset my-skill assets/icon.png          # Get skill asset

Standalone commands

bash
browser-cli status    # Check daemon + extension connection
browser-cli update    # Self-update to latest version

Workflow: Search, Interact, Verify

The recommended pattern for browser automation:

bash
# 1. Discover tabs
browser-cli tab list

# 2. Search for elements (fast, no screenshot needed)
browser-cli page search "{button,input,link}*" --tab 123

# 3. Interact using UIDs from search results
browser-cli interact click btn-submit --tab 123
# or
browser-cli interact fill input-email "user@example.com" --tab 123

# 4. Verify visually (only when needed)
browser-cli page screenshot

Login form example

bash
browser-cli page search "{input,textbox}*" --tab 123
browser-cli interact fill input-email "user@example.com" --tab 123
browser-cli interact fill input-pass "secret" --tab 123
browser-cli page search "*[Ll]ogin*" --tab 123
browser-cli interact click btn-login --tab 123

Shell script example

bash
#!/bin/bash
browser-cli tab new https://example.com
sleep 2
TAB_ID=$(browser-cli tab current | jq '.data.id')
browser-cli page search "link*" --tab "$TAB_ID"
browser-cli page screenshot

Global Options

OptionDefaultDescription
--port <n>9223Daemon WebSocket port
--host <h>127.0.0.1Daemon host address

Environment Variables

VariableDefaultDescription
BROWSER_CLI_WS_URLws://127.0.0.1:9223/cliOverride daemon WebSocket URL
BROWSER_CLI_CONNECT_TIMEOUT60000Connection timeout (ms)

Troubleshooting

SymptomFix
Daemon not runningRun any command to auto-spawn, or check with browser-cli status
Extension is not connectedOpen AIPex Options → WebSocket URL ws://localhost:9223/extension → Connect
Port 9223 in useUse --port 9224 and update extension URL
Timeout after 60sVerify extension is connected. Increase with BROWSER_CLI_CONNECT_TIMEOUT=120000
0 search resultsTry different patterns. Fall back to page screenshot --send-to-llm true + interact computer
Wrong resultsVerify tab ID with browser-cli tab list

相关 Skills

谷歌视频工具

by bwbernardweston18

热门

>

搜索与获取
未扫描4.5k

股票投研框架

by alondai

热门

股票投研9点分析框架，覆盖基本面/财务/竞品/估值/宏观/情绪等维度

搜索与获取
未扫描4.5k

SEO审计工具

by amdf01-debug

热门

搜索与获取
未扫描4.5k

相关 MCP 服务

Puppeteer 浏览器控制

编辑精选

by Anthropic

热门

Puppeteer 是让 Claude 自动操作浏览器进行网页抓取和测试的 MCP 服务器。

✎ 这个服务器解决了手动编写 Puppeteer 脚本的繁琐问题，适合需要自动化网页交互的开发者，比如抓取动态内容或做端到端测试。不过，作为参考实现，它可能缺少生产级的安全防护，建议在可控环境中使用。

搜索与获取
86.1k

Brave 搜索

编辑精选

by Anthropic

热门

Brave Search 是让 Claude 直接调用 Brave 搜索 API 获取实时网络信息的 MCP 服务器。

✎ 如果你想让 AI 助手帮你搜索最新资讯或技术文档，这个工具能绕过传统搜索的限制，直接返回结构化数据。特别适合需要实时信息的开发者，比如查 API 更新或竞品动态。不过它依赖 Brave 的 API 配额，高频使用可能受限。

搜索与获取
86.1k

网页抓取

编辑精选

by Anthropic

热门

Fetch 是 MCP 官方参考服务器，让 AI 能抓取网页并转为 Markdown 格式。

✎ 这个服务器解决了 AI 直接处理网页内容时格式混乱的问题，适合需要让 Claude 分析在线文档或新闻的开发者。不过作为参考实现，它缺乏生产级的安全配置，你得自己处理反爬虫和隐私风险。

搜索与获取
86.1k

评论