Leverages OpenAI models (e.g., GPT-4o) to power semantic element selection, self-healing, and AI analysis for browser automation tasks.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@AWebAgent navigate to google.com and search for 'AI agents'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AI 驱动的智能浏览器自动化 MCP 服务器，支持多 AI 提供商。
多 AI 提供商支持: OpenAI、阿里千问、DeepSeek、Moonshot、智谱 AI 等
语义选择器: 使用自然语言描述定位元素，无需 XPath/CSS
Self-Healing: 元素定位失败时自动诊断修复
HITL 中断: 关键操作前请求用户确认
Fast/Slow Path: 重复任务自动编译成低成本脚本
Related MCP server: Browser Automation MCP
# 安装依赖
npm install
# 安装浏览器
npm run playwright:install
# 配置 API 密钥
cp mcp-server/.env.example mcp-server/.env
# 编辑 .env 文件填入你的 API 密钥
# 运行
npm run mcp:dev变量 | 说明 | 示例 |
| API 密钥 |
|
| 提供商 |
|
| 模型名称 |
|
| 视觉模型 |
|
编辑 .cursor/mcp.json:
{
"mcpServers": {
"browser-automation": {
"command": "npx",
"args": ["tsx", "path/to/mcp-server/server.ts"],
"env": {
"AI_API_KEY": "your-api-key",
"AI_PROVIDER": "qwen"
}
}
}
}工具 | 说明 |
| 导航到 URL |
| 点击元素 |
| 输入文本 |
| 获取页面快照 |
| 智能执行复杂任务 |
| AI 分析页面 |
详细文档请查看 mcp-server/README.md
MIT
This server cannot be deployed
MCP server for Mint — AI-powered QA that runs your app in a real browser on every PR.
AI-powered browser automation — navigate, click, fill forms, and extract data from any website.
AI-powered web automation. Navigate websites using AI agents for one page or a thousand
AI-powered web automation. Navigate websites using AI agents for one page or a thousand
