A
alexzvency-ship-it/claude-browser-automator
Top Playwright Test Automation Framework 2026 - Run AI Browser Scripts Instantly
在 GitHub 查看 ↗链接已于 2026-11-12 核查 · HTTP 200 · 打开的是厂商自己的官网
安装与配置
客户端配置（取自该仓库的安装命令）
{
"mcpServers": {
"alexzvency-ship-it-claude-browser-automa": {
"command": "npx",
"args": ["auto-playwright"]
}
}
}解析自我们于 2026-11-07 抓取的该仓库 README；我们未执行过。
command/args 来自仓库写明的安装命令；不确定时请以该项目 README 为准。
VS Code（.vscode/mcp.json —— 键名是 servers）
{
"servers": {
"alexzvency-ship-it-claude-browser-automa": {
"command": "npx",
"args": ["auto-playwright"]
}
}
}VS Code 顶层键是 <code>servers</code>，其余与上面那份相同。
Claude Code（命令行）
claude mcp add --transport stdio alexzvency-ship-it-claude-browser-automa -- npx auto-playwright
Claude Code 不收 JSON 文件，用这条命令直接加。
配置项的键名以你的客户端为准：客户端与 CLI
- ★ 数取值
- 2 ★ · alexzvency-ship-it/claude-browser-automator · 取自 GitHub 搜索 API · （2026-11-13 20:00:50 核对）
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型：开源项目
- 收录与活跃
- 收录于 2026-11-08 · 最近见到 2026-11-14 · 共见到 7 次 · 被 1 个来源列出
- 评分构成
- 综合分 103 · 2 ★ · 1 个来源 · repo
- 安装
npx auto-playwright
常见问题（答案都来自我们的记录）
怎么安装？
该仓库给出的安装命令是
npx auto-playwright。我们的记录里它是仓库自己提供的；不保证在你的客户端上直接可用。它适合谁 / 用来做什么？
我们把它归在这些标签下: 开源项目。标签来自条目自身，不是我们强加的分类。
谁维护它、还在更新吗？
仓库
alexzvency-ship-it/claude-browser-automator；我们收录于 2026-11-08；最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。# AutoPlaywright: Dynamic Browser Automation Engine for AI-Assisted Testing
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
Version 1.0.0 | MIT License | 2026 Release
## 🌟 Why AutoPlaywright Exists
Imagine telling your browser exactly what to do—in plain English—and watching it comply. AutoPlaywright bridges the gap between human intent and machine execution by generating and running Playwright scripts dynamically through Claude AI. This isn't just another testing tool; it's a conversational browser companion that understands your testing goals and translates them into precise, executable code in real-time.
## 🔮 The Core Philosophy
Traditional browser automation requires you to think like a developer—writing selectors, waiting for elements, handling asynchronous operations. AutoPlaywright flips this paradigm: you describe the behavior, and the AI handles the syntax. It's the difference between being a mechanic building a car and a driver steering one.
## 🚀 Features That Redefine Browser Automation
### 🧠 AI-Powered Code Generation
Natural Language to Playwright: Describe actions like "Click the login button and wait for the dashboard" and watch Playwright code materialize
Contextual Intelligence: The AI understands page structure, element relationships, and typical user flows
Self-Healing Selectors: If an element changes, the system attempts to find alternatives automatically
### 🌐 Cross-Platform Compatibility
| OS | Chrome | Firefox | Safari | Edge |
| Windows | ✅ | ✅ | ❌ | ✅ |
| macOS | ✅ | ✅ | ✅ | ✅ |
| Linux | ✅ | ✅ | ❌ | ✅ |
| Docker | ✅ | ✅ | ❌ | ✅ |
### 📊 Visual Workflow Design
mermaid
graph TD
A[Natural Language Input] --> B[Claude AI Processing]
B --> C{Intent Analysis}
C --> D[Playwright Generation]
D --> E[Execution Engine]
E --> F[Result Verification]
F --> G[Success]
F --> H[Failure Analysis]
H --> B
G --> I[Report Generation]
### 🎯 Intelligent Execution Modes
Live Mode: Watch every action as it happens in a visible browser window
Headless Mode: Execute silently in the background for CI/CD pipelines
Hybrid Mode: Run critical paths visibly, non-critical paths in headless mode
## 💡 Example Profile Configuration
json
"project": "e-commerce-regression",
"ai-provider": "claude",
"api-key": "${CLAUDE_API_KEY}",
"browser": "chromium",
"viewport": { "width": 1920, "height": 1080 },
"timeout": 30000,
"retry-policy": {
"max-retries": 3,
"backoff-strategy": "exponential"
},
"reporting": {
"format": "html",
"include-screenshots": true,
"video-recording": "on-failure"
## 🎮 Example Console Invocation
bash
# Interactive mode - describe tests conversationally
npx auto-playwright --interactive
# Execute a saved test scenario
npx auto-playwright run --scenario checkout-flow.yaml
# Generate tests from Claude AI description
npx auto-playwright generate "Test the complete user registration flow including email verification"
# Continuous monitoring mode
npx auto-playwright monitor --url https://example.com --every 5m
## 🔧 Integration Ecosystem
### OpenAI API Integration
Leverage GPT-4 alongside Claude for enhanced test generation:
json
"provider": "openai",
"model": "gpt-4-turbo",
"temperature": 0.3,
"max-tokens": 2000
### Claude API Integration
The native integration offers specialized browser automation capabilities:
json
"provider": "claude",
"model": "claude-3-opus",
## 🌟 Why AutoPlaywright Exists
Imagine telling your browser exactly what to do—in plain English—and watching it comply. AutoPlaywright bridges the gap between human intent and machine execution by generating and running Playwright scripts dynamically through Claude AI. This isn't just another testing tool; it's a conversational browser companion that understands your testing goals and translates them into precise, executable code in real-time.
## 🔮 The Core Philosophy
Traditional browser automation requires you to think like a developer—writing selectors, waiting for elements, handling asynchronous operations. AutoPlaywright flips this paradigm: you describe the behavior, and the AI handles the syntax. It's the difference between being a mechanic building a car and a driver steering one.
## 🚀 Features That Redefine Browser Automation
### 🧠 AI-Powered Code Generation
Natural Language to Playwright: Describe actions like "Click the login button and wait for the dashboard" and watch Playwright code materialize
Contextual Intelligence: The AI understands page structure, element relationships, and typical user flows
Self-Healing Selectors: If an element changes, the system attempts to find alternatives automatically
### 🌐 Cross-Platform Compatibility
| OS | Chrome | Firefox | Safari | Edge |
| Windows | ✅ | ✅ | ❌ | ✅ |
| macOS | ✅ | ✅ | ✅ | ✅ |
| Linux | ✅ | ✅ | ❌ | ✅ |
| Docker | ✅ | ✅ | ❌ | ✅ |
### 📊 Visual Workflow Design
mermaid
graph TD
A[Natural Language Input] --> B[Claude AI Processing]
B --> C{Intent Analysis}
C --> D[Playwright Generation]
D --> E[Execution Engine]
E --> F[Result Verification]
F --> G[Success]
F --> H[Failure Analysis]
H --> B
G --> I[Report Generation]
### 🎯 Intelligent Execution Modes
Live Mode: Watch every action as it happens in a visible browser window
Headless Mode: Execute silently in the background for CI/CD pipelines
Hybrid Mode: Run critical paths visibly, non-critical paths in headless mode
## 💡 Example Profile Configuration
json
"project": "e-commerce-regression",
"ai-provider": "claude",
"api-key": "${CLAUDE_API_KEY}",
"browser": "chromium",
"viewport": { "width": 1920, "height": 1080 },
"timeout": 30000,
"retry-policy": {
"max-retries": 3,
"backoff-strategy": "exponential"
},
"reporting": {
"format": "html",
"include-screenshots": true,
"video-recording": "on-failure"
## 🎮 Example Console Invocation
bash
# Interactive mode - describe tests conversationally
npx auto-playwright --interactive
# Execute a saved test scenario
npx auto-playwright run --scenario checkout-flow.yaml
# Generate tests from Claude AI description
npx auto-playwright generate "Test the complete user registration flow including email verification"
# Continuous monitoring mode
npx auto-playwright monitor --url https://example.com --every 5m
## 🔧 Integration Ecosystem
### OpenAI API Integration
Leverage GPT-4 alongside Claude for enhanced test generation:
json
"provider": "openai",
"model": "gpt-4-turbo",
"temperature": 0.3,
"max-tokens": 2000
### Claude API Integration
The native integration offers specialized browser automation capabilities:
json
"provider": "claude",
"model": "claude-3-opus",
收录徽章（可选）
想把收录标记放进 README 就复制这行，可选，不影响收录。
[](https://anclum.com/zh/s/alexzvency-ship-it-claude-browser-automator/)
- 来源
- github · https://github.com/alexzvency-ship-it/claude-browser-automator
- 按领域浏览
- 开源项目
- Traction
- 2 ★
- 收录于
- 今天
同类推荐
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。