Mmcp.market

chrome-automation skill

by anbeime·anbeime/skill·7.3k stars

网页开发工程师与运营专员在需要自动化现有 Chrome 浏览器时必用!通过 CDP 协议无缝接管当前浏览器,自动配置环境并实时可视化执行任务。完美保留登录状态,一键完成复杂网页自动化流程,让繁琐操作自动化。

F10/100content scan

Is the chrome-automation skill safe?

A critical finding: do not install it without reading the flagged line. We read 13 files in the folder on 2026-09-28.

  • highSKILL.md:199

    Downloads a script and runs it in one step, so what runs is whatever that server sends that day. Common for installers, and still worth a look at the address.

    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
  • highscripts/auto-install-mac.sh:68

    Downloads a script and runs it in one step, so what runs is whatever that server sends that day. Common for installers, and still worth a look at the address.

    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --no-modify-path
  • highscripts/install-scripts.md:264

    Downloads a script and runs it in one step, so what runs is whatever that server sends that day. Common for installers, and still worth a look at the address.

    curl -fsSL https://raw.githubusercontent.com/vercel-labs/agent-browser/main/install.sh | bash

Install the chrome-automation skill

A skill is a folder. Copy it into your agent's skills folder and the agent loads it when the task matches its description. Read the findings above first.

git clone --depth 1 https://github.com/anbeime/skill.git /tmp/skill
mkdir -p ~/.claude/skills
cp -r /tmp/skill/skills/chrome-automation/chrome-automation ~/.claude/skills/chrome-automation
available in every project

In the Claude apps, zip the folder and upload it from the Skills settings. The folder on GitHub

The instructions your agent would load

SKILL.md as published, without the frontmatter. Read it on GitHub

Chrome Browser Automation with Real-time Control

Control the user's Google Chrome browser using agent-browser connected via Chrome DevTools Protocol (CDP). All actions are visible in real-time in the user's browser window.

⚠️ CRITICAL: Automatic Installation & Dependency Check

BEFORE using this skill, ALWAYS run the auto-installation check:

This skill automatically checks and installs all required dependencies without user confirmation. The installation process verifies:

  • agent-browser repository and binary
  • Node.js and pnpm
  • Playwright Chromium
  • Rust toolchain (cargo)
  • Native binary compilation

Auto-installation workflow:

# Mac: Run auto-install script
bash ~/.claude/skills/chrome-automation/scripts/auto-install-mac.sh

# Windows: Run auto-install script (PowerShell)
powershell -ExecutionPolicy Bypass -File "$HOME\.claude\skills\chrome-automation\scripts\auto-install-windows.ps1"

The script will:

  1. Check if agent-browser binary exists and works
  2. Install missing dependencies automatically (pnpm, Rust, etc.)
  3. Clone repository if not present
  4. Build native binary with proper Rust toolchain
  5. Verify installation success

Critical rules:

  • ❌ DO NOT fall back to using built-in browser tools (browserusedesktop)
  • ✅ MUST run auto-install script before any browser automation
  • ✅ Only proceed with browser automation AFTER successful installation
  • ✅ Script handles all dependency checks automatically

Why agent-browser is required:

  • Provides real-time visual feedback in user's actual Chrome browser
  • Allows user to see and interact with automation in progress
  • Offers precise element selection and interaction capabilities
  • Built-in browser tools are NOT a substitute for this skill's functionality

Quick Start

First-time setup (automatic)

Run the auto-install script first:

# Mac
bash ~/.claude/skills/chrome-automation/scripts/auto-install-mac.sh

# Windows (PowerShell)
powershell -ExecutionPolicy Bypass -File "$HOME\.claude\skills\chrome-automation\scripts\auto-install-windows.ps1"

The script automatically handles:

  • Repository cloning
  • Dependency installation (pnpm, Playwright)
  • Rust toolchain setup
  • Native binary compilation

Connect and control Chrome

IMPORTANT: Use the start-chrome script to preserve login states across sessions

# Mac: Start Chrome with persistent profile (RECOMMENDED)
# This preserves your login states, cookies, and settings
bash ~/.claude/skills/chrome-automation/scripts/start-chrome-mac.sh

# Windows: Start Chrome with persistent profile (PowerShell)
powershell -ExecutionPolicy Bypass -File "$HOME\.claude\skills\chrome-automation\scripts\start-chrome-windows.ps1"

# Use agent-browser with --cdp flag
cd /Users/jyxc-dz-0100272/Documents/agent-browser
AGENT_BROWSER_HOME=/Users/jyxc-dz-0100272/Documents/agent-browser \
./bin/agent-browser --cdp 9222 snapshot -i

What the start-chrome script does:

  1. First run: Creates a dedicated automation profile at ~/Library/Application Support/Google/Chrome-Automation
  2. First run: Imports your existing Chrome configuration (login states, cookies, settings)
  3. Subsequent runs: Uses the existing automation profile - all your logins are preserved
  4. You only need to log in once - it will be remembered for future sessions

Core Workflow

CRITICAL: Always show browser state to user after each significant action using screenshots.

  1. Connect: Ensure Chrome running with remote debugging port 9222
  2. Navigate: Open pages using --cdp 9222 flag
  3. Snapshot: Get interactive elements with snapshot -i
  4. Interact: Use element refs (@e1, @e2, etc.)
  5. Screenshot: Capture state and show to user
  6. Report: Describe what's visible in browser

Standard operation pattern

# Set path variables
export AB_HOME=/Users/jyxc-dz-0100272/Documents/agent-browser
cd $AB_HOME

# Navigate
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 open https://example.com
sleep 2

# Get interactive elements
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 snapshot -i

# Screenshot to show user
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 screenshot /tmp/state.png

# Interact
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 fill @e1 "text"
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 press Enter

# Capture result
sleep 2
AGENT_BROWSER_HOME=$AB_HOME ./bin/agent-browser --cdp 9222 screenshot /tmp/result.png

Installation & Setup

Automatic Installation (RECOMMENDED)

Always use the auto-install script - it handles all dependencies automatically:

Mac Installation

# Run auto-install script
bash ~/.claude/skills/chrome-automation/scripts/auto-install-mac.sh

What it does:

  1. Checks if agent-browser binary exists and works
  2. Clones repository if missing
  3. Installs Node.js dependencies (pnpm)
  4. Installs Playwright Chromium
  5. Checks/installs Rust toolchain (cargo, rustup)
  6. Sets up stable Rust toolchain if needed
  7. Builds native binary with npm run build:native
  8. Verifies installation success

Requirements:

  • Node.js (will notify if missing)
  • Git (for cloning repository)
  • Internet connection

Windows Installation

# Run auto-install script (PowerShell)
powershell -ExecutionPolicy Bypass -File "$HOME\.claude\skills\chrome-automation\scripts\auto-install-windows.ps1"

What it does:

  1. Checks if agent-browser binary exists and works
  2. Clones repository if missing
  3. Installs Node.js dependencies (pnpm)
  4. Installs Playwright Chromium
  5. Downloads and installs Rust toolchain if needed
  6. Sets up stable Rust toolchain
  7. Builds native binary with npm run build:native
  8. Verifies installation success

Requirements:

  • Node.js (will notify if missing - download from https://nodejs.org/)
  • Git (for cloning repository)
  • Internet connection

Manual Installation (Advanced Users Only)

If you prefer manual installation or need to troubleshoot:

Mac Manual Steps

# 1. Clone repository
cd ~/Documents
git clone https://github.com/vercel-labs/agent-browser.git
cd agent-browser

# 2. Install Node dependencies
pnpm install
npx playwright install chromium

# 3. Install Rust (if not installed)
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
export PATH="$HOME/.cargo/bin:$PATH"
rustup default stable

# 4. Build native binary
npm run build:native

Windows Manual Steps

# 1. Clone repository
Set-Location "$HOME\Documents"
git clone https://github.com/vercel-labs/agent-browser.git
Set-Location agent-browser

# 2. Install Node dependencies
pnpm install
npx playwright install chromium

# 3. Install Rust (if not installed)
# Download from: https://win.rustup.rs/x86_64
# Or use the auto-install script

# 4. Build native binary
npm run build:native

Verify Installation

After installation (automatic or manual), verify it works:

# Mac
cd ~/Documents/agent-browser
AGENT_BROWSER_HOME=~/Documents/agent-browser ./bin/agent-browser --version

# Windows (PowerShell)
Set-Location "$HOME\Documents\agent-browser"
$env:AGENT_BROWSER_HOME = "$HOME\Documents\agent-browser"
.\bin\agent-browser.cmd --version

Troubleshooting Installation

"No binary found for darwin-arm64" or "No binary found for win32-x64":

  • Rust is not installed or not in PATH
  • Run the auto-install script, it will handle Rust installation
  • Or manually install Rust and run npm run build:native

"cargo: command not found":

  • Rust toolchain not in PATH
  • Mac: export PATH="$HOME/.cargo/bin:$PATH"
  • Windows: $env:PATH = "$HOME\.cargo\bin;$env:PATH"
  • Then run rustup default stable and npm run build:native

"rustup could not choose a version":

More skills from anbeime/skill

  • APDF Processing Pro综合办公文员与软件开发工程师当需要批量处理PDF表单、提取表格或进行OCR识别时,使用内置脚本一键完成自动化提取与数据校验,彻底告别繁琐的手动录入,让复杂文档工作流高效、稳健落地。
  • Aagent-team产品经理与项目管理专家在应对复杂项目时,使用此技能可动态组建包含“执行、指挥、评审”的专属AI团队。实时查看多智能体辩论与决策全过程,共享统一上下文记忆,一键完成从会议决策到系统构建的全流程高效协同!
  • Aagentkit-multimedia-shopping电商运营与内容创作者在需要制作带货短视频时,用此技能一键生成9:16竖屏数字人成片。自动编排AI绘画、语音合成与视频生成,快速产出“小省导购员”专属形象与专业配音,让多模态视频制作省时省力。
  • Aantinet-doc-parse软件开发工程师与数据科学家在构建RAG系统时,当需处理PDF/Word/Excel等多格式复杂文档,用此技能可自动触发三级解析降级,一键输出高置信度结构化Markdown与元数据,免去繁琐清洗,直接夯实企业知识库数据底座!
  • Aantinet-four-color-cards财务和投资分析师、律师助理与法律助手在撰写金融研报或分析法律文档时,必用此技能!基于解析文档一键生成事实蓝卡、解释绿卡、风险黄卡与行动红卡。自动完成深度剖析与风险排查,产出100%可溯源的结构化四色报告,让复杂分析高效且严谨。
  • Aantinet-provenance软件开发工程师与信息安全分析师在构建多智能体系统时,当需全链路可观测与安全审计,请用此技能。它自动收集操作日志,生成可追溯证据链与向量索引,开箱即用实现系统审计与精准回放,让Agent运行安全透明。
  • Aantinet-security-scan信息安全分析师与软件开发工程师在搭建文档处理流水线时,当需要拦截违规文件或外部URL,请挂载此技能作为强制前置安检,自动输出 pass/reject 判定与详细合规扫描报告,一键守住零信任安全底线。
  • AarchifyCreate polished, validated architecture, workflow, sequence, data-flow, and lifecycle/state diagrams as explorable standalone HTML with inline SVG, dark/light themes, optional trace motion, and PNG/JPEG/WebP/SVG/WebM export. Accept plain-language requirements or pasted Mermaid flowchart, sequenceDiagram, and stateDiagram input; inspect repository evidence when the diagram must reflect real code. Use when the user asks to visualize system architecture, infrastructure, cloud/security/network topology, technical workflows, API call sequences, request lifecycles, data pipelines, ETL/ELT, data lineage, state machines, or to convert/beautify Mermaid.
  • Aarticle-illustrator内容创作者与自媒体创作者在撰写长文缺乏视觉辅助时,当需要给文章配图或生成插图请使用。自动分析文章结构,精准识别位置并生成插画插入段落,将抽象概念具象化。一键完成智能图文排版,省去手动找图时间,大幅提升阅读体验!
  • Aarticle-illustrator内容创作者和自媒体创作者在撰写长文时,当需要“给文章配图”或“添加插图”时使用。自动分析文章结构,精准识别需要视觉辅助的位置,生成并插入契合语境的插画,将抽象概念具象化,一键完成高质量图文排版,大幅提升阅读体验。
  • Aatutun-xhs-cover阿囤囤风格小红书封面提示词生成。用于AI工具、效率方法、产品测评、教程部署、个人经验、播客推荐、职场认知类内容的爆款竖版封面设计。特点是真人出镜、超大柔和浅黄(#FDFFA7)/白色中文标题、粗黑描边、人物抠图描边、绿色勾选清单、emoji贴纸、箭头标注、高对比高密度构图。关键词:小红书封面、爆款封面、阿囤囤风格、AI教程封面、产品测评封面、封面提示词。
  • Abaoyu-format-markdown技术写作员与内容创作者在处理纯文本或 Markdown 稿件时,若需优化文章结构、添加 frontmatter 或自动修正中英文排版间距,使用此技能可一键生成专业美观的 {filename}-formatted.md 文件,让文档排版更完美。

All agent skills → · MCP servers