Bright Data MCP

Bright Data is one of the biggest web data companies, and its official MCP server gives Claude its scraping muscle. Claude can search the web, read pages that normally block bots and run ready-made scrapers that return clean data from Amazon, LinkedIn, Instagram, TikTok, Reddit and many more.

For marketers it's handy for competitor and pricing research, lead lists and social listening. It can also put a prompt to ChatGPT or Perplexity and return the answer, which is a quick way to check how AI assistants talk about your brand.

It's hosted, so setup is one command with your API token from Bright Data's account settings. The free tier covers 5,000 requests a month. After that you pay as you go, and you can set a spending cap. All 32 tools together take up a lot of context, so use the groups option to load only what you need.

• Search Google, Bing or Yandex live
• Read any web page as clean text, even past CAPTCHAs and bot blockers
• Pull structured data from Amazon, LinkedIn, TikTok, Instagram, Reddit and more
• Drive a remote browser to click, type, scroll and take screenshots
• Ask ChatGPT or Perplexity a question and see how they answer it

Open your terminal in the project folder and run:

claude mcp add --transport http brightdata "https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN"

Other ways to set it up:

# Load fewer tools to save context by adding groups to the URL, e.g.
https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN&groups=scrapers,browser

You'll be asked to sign in or paste an API key the first time you use it.

Marketers, researchers and founders who need data from websites at scale.

• Competitor and pricing research
• Scraping social media profiles
• Lead generation
• Checking brand visibility in AI answers