Skip to main content
Glama
README.md
<div align="center">
  <a href="https://www.bright.cn/ai/mcp-server">
    <img src="https://github.com/user-attachments/assets/c21b3f7b-7ff1-40c3-b3d8-66706913d62f" alt="Bright Data Logo">
  </a>

  <h1>The Web MCP</h1>
  
  <p>
    <strong>🌐 让你的 AI 拥有实时上网超能力</strong><br/>
    <i>无缝连接 LLM 到实时网络,不再被封锁</i>
  </p>

  <p>
    <a href="https://www.npmjs.com/package/@brightdata/mcp">
      <img src="https://img.shields.io/npm/v/@brightdata/mcp?style=for-the-badge&color=blue" alt="npm version"/>
    </a>
    <a href="https://www.npmjs.com/package/@brightdata/mcp">
      <img src="https://img.shields.io/npm/dw/@brightdata/mcp?style=for-the-badge&color=green" alt="npm downloads"/>
    </a>
    <a href="https://github.com/bright-cn/brightdata-mcp/blob/main/LICENSE">
      <img src="https://img.shields.io/badge/license-MIT-purple?style=for-the-badge" alt="License"/>
    </a>
  </p>

  <p>
    <a href="#-quick-start">快速开始</a> •
    <a href="#-features">功能特性</a> •
    <a href="#-pricing--modes">价格与模式</a> •
    <a href="#-demos">演示</a> •
    <a href="#-documentation">文档</a> •
    <a href="#-support">支持</a>
  </p>

  <div>
    <h3>🎉 <strong>提供免费套餐!</strong> 🎉</h3>
    <p><strong>每月 5,000 次请求免费</strong> <br/>
    <sub>非常适合原型开发与日常 AI 工作流</sub></p>
  </div>
</div>

---

## 🌟 概览

The Web MCP 是为 AI 助手赋予真正「上网能力」的通道。告别过时回答,也不再出现“我无法访问实时信息”的窘境——只需稳定、顺畅、可用的网络访问。

由全球领先的[Bright Data](https://www.bright.cn) Web 数据平台打造,这个 MCP 服务器确保你的 AI 不再被封锁、限流或被验证码拦住。

<div align="center">
  <table>
    <tr>
      <td align="center">✅ <strong>适配任意 LLM</strong><br/><sub>Claude、GPT、Gemini、Llama</sub></td>
      <td align="center">🛡️ <strong>不再被封</strong><br/><sub>企业级解封与反封锁</sub></td>
      <td align="center">🚀 <strong>5,000 次/月免费</strong><br/><sub>按月计</sub></td>
      <td align="center">⚡ <strong>零配置</strong><br/><sub>开箱即用</sub></td>
    </tr>
  </table>
</div>

---

## 🎯 适用场景

- 🔍 实时调研——获取最新价格、新闻与实时数据
- 🛍️ 电商情报——监测产品、价格与库存  
- 📊 市场分析——跟踪竞品与行业趋势
- 🤖 AI 代理——构建真正能「浏览网页」的智能体
- 📝 内容创作——为写作获取最新可信的信息
- 🎓 学术研究——高效汇聚多源公开数据

---

## ⚡ 快速开始


<summary><b>📡 使用托管服务器——无需安装!</b></summary>

零门槛使用,只需在你的 MCP 客户端中添加以下 URL:

```
https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN_HERE
```

在 Claude Desktop 中设置:
1. 打开:Settings → Connectors → Add custom connector
2. 名称:`Bright Data Web`
3. URL:`https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN`
4. 点击 “Add”,完成!✨


<summary><b>在本地运行</b></summary>

```json
{
  "mcpServers": {
    "Bright Data": {
      "command": "npx",
      "args": ["@brightdata/mcp"],
      "env": {
        "API_TOKEN": "<your-api-token-here>"
      }
    }
  }
}
```

---

## 🚀 价格与模式

<div align="center">
  <table>
    <tr>
      <th width="33%">⚡ Rapid 模式(免费)</th>
      <th width="33%">💎 Pro 模式</th>
    </tr>
    <tr>
      <td align="center">
        <h3>$0/月</h3>
        <p><strong>5,000 次请求</strong></p>
        <hr/>
        <p>✅ Web 搜索<br/>
        ✅ 使用 Web Unlocker 抓取<br/>
        ❌ 浏览器自动化<br/>
        ❌ Web 数据工具</p>
        <br/>
        <code>默认模式</code>
      </td>
      <td align="center">
        <h3>按量计费</h3>
        <p><strong>包含 Rapid 全部内容 + 60+ 高级工具</strong></p>
        <hr/>
        <p>✅ 浏览器控制<br/>
        ✅ Web 数据 API<br/>
        <br/>
        <br/>
        <br/>
        <code>PRO_MODE=true</code>
      </td>
    </tr>
  </table>
</div>

> 💡 注意:Pro 模式不包含在免费套餐内,按使用量额外计费。

---

## ✨ 功能特性

### 🔥 核心能力

<table>
  <tr>
    <td>🔍 <b>智能网页搜索</b><br/>面向 AI 优化的类 Google 级结果</td>
    <td>📄 <b>干净 Markdown</b><br/>适配 AI 的内容抽取输出</td>
  </tr>
  <tr>
    <td>🌍 <b>全球可达</b><br/>自动绕过地理限制</td>
    <td>🛡️ <b>反机器人与防封</b><br/>不再被封、不再限流</td>
  </tr>
  <tr>
    <td>🤖 <b>浏览器自动化</b><br/>远程控制真实浏览器(Pro)</td>
    <td>⚡ <b>低时延</b><br/>为快速响应优化</td>
  </tr>
</table>

### 🎯 一些“开箱即用”的查询示例

```yaml
✅ “特斯拉当前股价是多少?”
✅ “现在东京评分最高的餐厅有哪些?”
✅ “纽约今天的天气预报是什么?”
✅ “本周有哪些电影上映?”
✅ “今天 Twitter 上的热点话题是什么?”
```

---

## 🎬 演示

> 注:以下视频展示的是早期版本,新的演示即将上线!🎥

<details>
<summary><b>查看演示视频</b></summary>

### 基础网页搜索演示
https://github.com/user-attachments/assets/59f6ebba-801a-49ab-8278-1b2120912e33

### 高级抓取演示
https://github.com/user-attachments/assets/61ab0bee-fdfa-4d50-b0de-5fab96b4b91d

[📺 更多教程(YouTube)→](https://github.com/bright-cn/brightdata-mcp/blob/main/examples/README.md)

</details>

---

## 🔧 可用工具

### ⚡ Rapid 模式工具(默认,免费)

| 工具 | 描述 | 适用场景 |
|------|------|----------|
| 🔍 `search_engine` | 面向 AI 优化的 Web 搜索 | 研究、查证、时事 |
| 📄 `scrape_as_markdown` | 将任意网页转为干净的 Markdown | 内容抽取、文档化 |

### 💎 Pro 模式工具(60+)

<details>
<summary><b>点击查看全部 Pro 工具</b></summary>

| 类别 | 工具 | 描述 |
|------|------|------|
| 浏览器控制 | `scraping_browser.*` | 全功能浏览器自动化 |
| Web 数据 API | `web_data_*` | 结构化数据抽取 |
| 电商 | 产品抓取器 | Amazon、eBay、Walmart 等 |
| 社媒 | 社交抓取器 | Twitter、LinkedIn、Instagram |
| 地图与本地 | 位置工具 | Google 地图、商家数据 |

[📚 查看完整工具文档 →](https://github.com/bright-cn/brightdata-mcp/blob/main/assets/Tools.md)

</details>

---

## 🎮 立即试用!

### 🧪 在线 Playground
无需任何配置,直接体验 Web MCP:

<div align="center">
  <a href="https://www.bright.cn/ai/playground-chat">
    <img src="https://img.shields.io/badge/Try_on-Playground-00C7B7?style=for-the-badge&logo=data:image/svg+xml;base64,PHN2ZyB3aWR0aD0iMjQiIGhlaWdodD0iMjQiIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KPHBhdGggZD0iTTEyIDJMMyA3VjE3TDEyIDIyTDIxIDE3VjdMMTIgMloiIHN0cm9rZT0id2hpdGUiIHN0cm9rZS13aWR0aD0iMiIvPgo8L3N2Zz4=" alt="Playground"/>
  </a>
</div>

---

## 🔧 配置

### 基础配置
```json
{
  "mcpServers": {
    "Bright Data": {
      "command": "npx",
      "args": ["@brightdata/mcp"],
      "env": {
        "API_TOKEN": "your-token-here"
      }
    }
  }
}
```

### 高级配置
```json
{
  "mcpServers": {
    "Bright Data": {
      "command": "npx",
      "args": ["@brightdata/mcp"],
      "env": {
        "API_TOKEN": "your-token-here",
        "PRO_MODE": "true",              // 启用 60+ 全部工具
        "RATE_LIMIT": "100/1h",          // 自定义限速
        "WEB_UNLOCKER_ZONE": "custom",   // 自定义解封 Zone
        "BROWSER_ZONE": "custom_browser" // 自定义浏览器 Zone
      }
    }
  }
}
```

---

## 📚 文档

<div align="center">
  <table>
    <tr>
      <td align="center">
        <a href="https://docs.brightdata.com/mcp-server/overview">
          <img src="https://img.shields.io/badge/📖-API_Docs-blue?style=for-the-badge" alt="API Docs"/>
        </a>
      </td>
      <td align="center">
        <a href="https://github.com/bright-cn/brightdata-mcp/blob/main/examples">
          <img src="https://img.shields.io/badge/💡-Examples-green?style=for-the-badge" alt="Examples"/>
        </a>
      </td>
      <td align="center">
        <a href="https://github.com/bright-cn/brightdata-mcp/blob/main/CHANGELOG.md">
          <img src="https://img.shields.io/badge/📝-Changelog-orange?style=for-the-badge" alt="Changelog"/>
        </a>
      </td>
      <td align="center">
        <a href="https://www.bright.cn/blog/ai/web-scraping-with-mcp">
          <img src="https://img.shields.io/badge/📚-Tutorial-purple?style=for-the-badge" alt="Tutorial"/>
        </a>
      </td>
    </tr>
  </table>
</div>

---

## 🚨 常见问题与解决方案

<details>
<summary><b>🔧 故障排查指南</b></summary>

### ❌ “spawn npx ENOENT” 错误
解决:安装 Node.js,或使用 node 的完整路径:
```json
"command": "/usr/local/bin/node"  // macOS/Linux
"command": "C:\\Program Files\\nodejs\\node.exe"  // Windows
```

### ⏱️ 复杂站点超时
解决:在客户端设置中将超时提高到 180s

### 🔑 鉴权问题
解决:确认 API Token 有效且具备相应权限

### 📡 远程连接问题
解决:检查网络与防火墙设置

[更多排查 →](https://github.com/bright-cn/brightdata-mcp#troubleshooting)

</details>

---

## 🤝 参与贡献

我们欢迎各种形式的贡献:

- 🐛 [提交 Bug](https://github.com/bright-cn/brightdata-mcp/issues)
- 💡 [功能建议](https://github.com/bright-cn/brightdata-mcp/issues)
- 🔧 [提交 PR](https://github.com/bright-cn/brightdata-mcp/pulls)
- ⭐ 欢迎为本仓库加星!

请遵循 [Bright Data 的代码规范](https://www.bright.cn/dna/js_code)。

---

## 📞 支持

<div align="center">
  <table>
    <tr>
      <td align="center">
        <a href="https://github.com/bright-cn/brightdata-mcp/issues">
          <strong>🐛 GitHub Issues</strong><br/>
          <sub>提交问题与需求</sub>
        </a>
      </td>
      <td align="center">
        <a href="https://docs.brightdata.com/mcp-server/overview">
          <strong>📚 文档</strong><br/>
          <sub>完整指南</sub>
        </a>
      </td>
      <td align="center">
        <a href="mailto:support@brightdata.com">
          <strong>✉️ 邮件</strong><br/>
          <sub>support@brightdata.com</sub>
        </a>
      </td>
    </tr>
  </table>
</div>

---

## 📜 许可证

MIT © [Bright Data Ltd.](https://www.bright.cn)

---

<div align="center">
  <p>
    <strong>由 ❤️ 驱动</strong><br/>
    <a href="https://www.bright.cn">
      <img src="https://idsai.net.technion.ac.il/files/2022/01/Logo-600.png" alt="Bright Data" height="30"/>
    </a>
  </p>
  <p>
    <sub>全球第一的 Web 数据平台</sub>
  </p>
  
  <br/>
  
  <p>
    <a href="https://github.com/bright-cn/brightdata-mcp">⭐ 欢迎为本仓库加星</a> • 
    <a href="https://www.bright.cn/blog">阅读我们的博客</a>
  </p>
</div>

TDQS

A3.6/5.0

Scored across 4 tools

Disambiguation5/5

Each tool has a clearly distinct purpose: scrape_as_markdown handles single webpage scraping, scrape_batch handles multiple webpages, search_engine scrapes single search queries, and search_engine_batch handles multiple search queries. There is no overlap in functionality, making tool selection straightforward for an agent.

Naming Consistency5/5

All tool names follow a consistent snake_case pattern with clear verb_noun structures: scrape_as_markdown, scrape_batch, search_engine, and search_engine_batch. The naming is predictable and aligns well with the actions each tool performs.

Tool Count5/5

With 4 tools, the server is well-scoped for web scraping and search functionality. Each tool earns its place by covering distinct aspects of the domain: single and batch operations for both webpage scraping and search engine queries, avoiding bloat while providing essential coverage.

Completeness4/5

The tool set covers core web scraping and search operations effectively, including single and batch modes. A minor gap exists in lacking tools for more advanced interactions like form submission or dynamic content handling, but the provided tools support key workflows without dead ends.

Maintenance

ActivityInactive
ResponsivenessNo issues