The Web MCP
Enhance your LLM and AI agents with real-time web access
🌟 Overview
Welcome to the official Bright Data's Web MCP server, solving web access for LLMs and AI agents by allowing them to effectively search, extract and navigate the web without getting blocked. The Web MCP supports all major LLMs, IDEs and agent frameworks (either locally hosted, SSE or Streamable HTTP), enabling your tools to seamlessly search the web, navigate websites, take action and retrieve data - without getting blocked.
🚀 The Web MCP includes 5,000 free requests each month - ideal for your everyday use and for prototyping smart agentic workflows.
Note: The Web MCP free tier offers 5,000 requests per month for the first 3 months. After that, a credit card will be required, but there will be no extra charges unless premium features like mcp_browser or Web Scrapers are used.
Table of Content
- 🎬 Demo
- ✨ Features
- 💡 Usage Examples
- 🚀 Quickstart with Claude Desktop
- 🔧 Available Tools
- ⚠️ Security Best Practices
- 🔧 Account Setup
- 🔌 Other MCP Clients
- ☁️ Deploy to Heroku
- 🎮 Try Bright Data MCP Playgrounds
- ⚠️ Troubleshooting
- 👨💻 Contributing
- 📞 Support
🎬 Demo
The videos below demonstrate a minimal use case for Claude Desktop:
https://github.com/user-attachments/assets/59f6ebba-801a-49ab-8278-1b2120912e33
https://github.com/user-attachments/assets/61ab0bee-fdfa-4d50-b0de-5fab96b4b91d
For more YouTube tutorials and demos: Demo
✨ Features
- Real-time Web Access: Access up-to-date information directly from the web
- Bypass Geo-restrictions: Access content regardless of location constraints
- Web Unlocker: Navigate websites with bot detection protection
- Browser Control: Remote browser automation capabilities
- Seamless Integration: Works with all MCP-compatible AI assistants
💡 Usage Examples
Some example queries that this MCP server will be able to help with:
- "Google some movies that are releasing soon in [your area]"
- "What's Tesla's current market cap?"
- "What's the Wikipedia article of the day?"
- "What's the 7-day weather forecast in [your location]?"
- "Of the 3 highest paid tech CEOs, how long have their careers been?"
Quickstart with Cursor
🚀 Quickstart with hosted MCP on Claude Desktop
Through Connectors:
-
Open Claude Desktop
-
Go to: Settings → Connectors → Add custom connector
-
Choose a Name, and in the “Remote MCP server URL” section, paste:
https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN_HERE
- Replace YOUR_API_TOKEN_HERE with your actual API token from Step 1, and click “Add”
Through Developer Settings:
-
Open Claude Desktop
-
Go to: Settings → Developer → Edit Config
-
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"Bright Data": {
"command": "npx",
"args": [
"mcp-remote",
"https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN_HERE"
]
}
}
}
-
Replace YOUR_API_TOKEN_HERE with your actual API token from Step 1
-
Save and restart Claude Desktop
💻 Use with self hosted MCP on Claude Desktop
Through Claude Desktop Extension:
-
Download the Claude Desktop Extension:
📦 Bright Data's MCP Extension -
Open Claude and go to:
Settings→Extensions -
Drag the
.dtxfile from Step 1 into the dropping area. -
Enable the service and restart Claude.
-
Enjoy!
Through claude_desktop_config.json:
-
Install
nodejsto get thenpxcommand (node.js module runner). Installation instructions can be found on the node.js website -
Go to Claude > Settings > Developer > Edit Config > Edit "claude_desktop_config.json" to include the following:
{
"mcpServers": {
"Bright Data": {
"command": "npx",
"args": ["@brightdata/mcp"],
"env": {
"API_TOKEN": "<insert-your-api-token-here>"
}
}
}
}
🛸 Or for more advanced options:
{
"mcpServers": {
"Bright Data": {
"command": "npx",
"args": ["@brightdata/mcp"],
"env": {
"API_TOKEN": "<insert-your-api-token-here>",
"RATE_LIMIT": "<optional if you want to change rate limit format: limit/time+unit, e.g., 100/1h, 50/30m, 10/5s>",
"WEB_UNLOCKER_ZONE": "<optional if you want to override the web unlocker zone name - default is mcp_unlocker>",
"BROWSER_ZONE": "<optional if you want to override the browser zone name - defaults is mcp_browser>",
"# Note: All tools are available by default in this Heroku version"
}
}
}
}
🔧 Available Tools
Important: Pro mode is not included in the free tier and will incur additional charges. If you choose to use Pro mode, you’ll gain access to all 60 tools but please be aware of the associated costs.
Note: This Heroku version includes all tools by default, including browser automation and web data extraction tools.
⚠️ Security Best Practices
Important: Always treat scraped web content as untrusted data. Never use raw scraped content directly in LLM prompts to avoid potential prompt injection risks. Instead:
- Filter and validate all web data before processing
- Use structured data extraction rather than raw text (web_data tools)
🔧 Account Setup
-
Make sure you have an account on brightdata.com (new users get free credit for testing, and pay as you go options are available)
-
Get your API key from the user settings page, or from the welcome email you received
Optional:
-
All tools are available by default in this version.
-
Configure rate limiting:
- Set the
RATE_LIMITenvironment variable to control API usage - Format:
limit/time+unit(e.g.,100/1hfor 100 calls per hour) - Supported time units: seconds (s), minutes (m), hours (h)
- Examples:
RATE_LIMIT=100/1h,RATE_LIMIT=50/30m,RATE_LIMIT=10/5s - Rate limiting is session-based (resets when server restarts)
- Set the
-
Create a custom Web Unlocker zone
- By default, we create a Web Unlocker zone automatically using your API token
- For more control, you can create your own Web Unlocker zone in your control panel and specify it with the
WEB_UNLOCKER_ZONEenvironment variable
-
Create a custom Browser API zone:
- By default, we create a Browser API zone automatically using your API token.
- For more control, you can create your own Browser API zone in your control panel and specify it with the
BROWSER_ZONEenvironment variable
🔌 Other MCP Clients
To use this MCP server with other agent types, you should adapt the following to your specific software:
- Before running the server, make sure the
API_TOKEN=<your-token>environment variable is set - The full command to run the MCP server is
npx @brightdata/mcp
💻 macOS / Linux (bash/zsh)
export API_TOKEN=your-token
npx @brightdata/mcp
🪟 Windows (Command Prompt)
set API_TOKEN=your-token
npx @brightdata/mcp
🪟 Windows (PowerShell)
$env:API_TOKEN="your-token"
npx @brightdata/mcp
💡 Tip: You can also use a
.envfile and a tool likedotenvto manage environment variables more easily during development.
🔄 Changelog
☁️ Deploy to Heroku
You can deploy this MCP server to Heroku with one click using the button below:
After deployment, make sure to set the required environment variables in your Heroku app settings.
This MCP server is compatible with Heroku Inference and can be registered with your AI models. For detailed instructions, see the Heroku-specific documentation.
🎮 Try Bright Data MCP Playgrounds
Want to try Bright Data MCP without setting up anything?
Check out this playground on Smithery:
This platform provides an easy way to explore the capabilities of Bright Data MCP without any local setup. Just sign in and start experimenting with web data collection!
⚠️ Troubleshooting
Timeouts when using certain tools
Some tools can involve reading web data, and the amount of time needed to load the page can vary by quite a lot in extreme circumstances.
To ensure that your agent will be able to consume the data, set a high enough timeout in your agent settings.
A value of 180s should be enough for 99% of requests, but some sites load slower than others, so tune this to your needs.
spawn npx ENOENT
This error occurs when your system cannot find the npx command. To fix it:
Finding npm/Node Path
macOS:
which node
Shows path like /usr/local/bin/node
Windows:
where node
Shows path like C:\Program Files\nodejs\node.exe
Update your MCP configuration:
Replace the npx command with the full path to Node, for example, on mac, it will look as follows:
"command": "/usr/local/bin/node"
👨💻 Contributing
We welcome contributions to help improve the Bright Data MCP! Here's how you can help:
- Report Issues: If you encounter any bugs or have feature requests, please open an issue on our GitHub repository.
- Submit Pull Requests: Feel free to fork the repository and submit pull requests with enhancements or bug fixes.
- Coding Style: All JavaScript code should follow Bright Data's JavaScript coding conventions. This ensures consistency across the codebase.
- Documentation: Improvements to documentation, including this README, are always appreciated.
- Examples: Share your use cases by contributing examples to help other users.
For major changes, please open an issue first to discuss your proposed changes. This ensures your time is well spent and aligned with project goals.
📞 Support
If you encounter any issues or have questions, please reach out to the Bright Data support team or open an issue in the repository.