Bright Data MCP

Installation & Setup

Quick Installation with NPX

npx @brightdata/mcp

Environment Variables

• API_TOKEN: Your Bright Data API token (required)
• WEB_UNLOCKER_ZONE: Custom zone name (optional, defaults to 'mcp_unlocker')
• BROWSER_AUTH: Browser authentication for browser control tools (optional)

Configuring MCP Clients

Claude Desktop Configuration

1. Edit the Claude Desktop configuration file through VS code:

   # macOS
   code "$HOME/Library/Application Support/Claude/claude_desktop_config.json"
   
   # Windows
   code %APPDATA%\Claude\claude_desktop_config.json
   

2. Add the Bright Data server configuration:

   {
     "mcpServers": {
       "Bright Data": {
         "command": "npx",
         "args": ["@brightdata/mcp"],
         "env": {
           "API_TOKEN": "<insert-your-api-token-here>",
           "WEB_UNLOCKER_ZONE": "<optional-zone-name>",
           "BROWSER_AUTH": "<optional-browser-auth>"
         }
       }
     }
   }
   

3. Save the file and restart Claude Desktop.

Other MCP Clients (Cursor, Windsurf, etc.)

Configure your MCP client to run the command npx @brightdata/mcp with the required environment variables.

Available Tools

Web Search & Scraping

• search_engine: Search Google, Bing, or Yandex and get results in markdown format
• scrape_as_markdown: Extract webpage content as Markdown
• scrape_as_html: Extract webpage content as HTML
• session_stats: View tool usage statistics for the current session

Structured Data Extraction

• web_data_amazon_product: Extract structured Amazon product data
• web_data_amazon_product_reviews: Extract Amazon product reviews
• web_data_linkedin_person_profile: Extract LinkedIn person profiles
• web_data_linkedin_company_profile: Extract LinkedIn company profiles

Browser Control Tools (with BROWSER_AUTH)

• scraping_browser_navigate: Navigate to a URL
• scraping_browser_go_back: Go back to the previous page
• scraping_browser_go_forward: Go forward to the next page
• scraping_browser_links: Get all links on the current page
• scraping_browser_click: Click on an element
• scraping_browser_type: Type text into an element
• scraping_browser_wait_for: Wait for an element to appear
• scraping_browser_screenshot: Take a screenshot
• scraping_browser_get_html: Get the HTML content of the page
• scraping_browser_get_text: Get the text content of the page

Example Usage

Web Search Examples

Search for recent news about artificial intelligence

Web Scraping Examples

Scrape the content from https://example.com/article as markdown

Structured Data Examples

Extract product information from https://www.amazon.com/dp/B08L5TNJHG

Browser Control Examples

Navigate to wikipedia.org, search for "machine learning", and take a screenshot of the results

Account Setup

1. Create an account on brightdata.com (new users get free credit)
2. Get your API key from the user settings page
3. Create a Web Unlocker proxy zone called mcp_unlocker in your control panel
4. (Optional) For browser control tools:
   • Create a "scraping browser" zone in your Brightdata control panel
   • Copy the authentication string from the Scraping Browser overview tab

Troubleshooting

Timeouts When Using Tools

• Some web pages may take longer to load than others
• Set a high enough timeout in your agent settings (180s recommended)
• Adjust based on your specific needs and the websites you're accessing

Connection Issues

• Verify your API token is correct
• Ensure your Web Unlocker zone is properly configured
• Check that your Scraping Browser authentication is valid (if using browser tools)

Browser Tools Not Available

• Confirm that the BROWSER_AUTH environment variable is set correctly
• Verify that your Scraping Browser zone is active in the Bright Data control panel