Bright Data MCP server for AI apps

What your AI apps can do with Bright Data

Search engines and scrape pages
Extract structured web data

Bright Data MCP tools

32 tools, taken from the publisher's documentation on 29 October 2026. The server decides what it offers and can add or rename tools at any time; after you connect, Toolstile shows the current list.

Tool	What it does
`search_engine`	Scrape search results from Google, Bing or Yandex
`scrape_as_markdown`	Scrape a single webpage and return Markdown, handling bot protection
`search_engine_batch`	Run up to 10 search queries in parallel
`scrape_batch`	Scrape up to 10 webpages in one request as Markdown
`discover`	Search the web and rank results by AI relevance to an intent
`scrape_as_html`	Scrape a single webpage and return the HTML body
`extract`	Scrape a webpage and convert it to structured JSON using AI
`session_stats`	Report how many times each tool was called in the current session
`search_scrapers`	Find a pre-built scraper by site or data type
`get_scraper_details`	Show a scraper's description, input fields and sample input
`run_scraper`	Run a pre-built scraper on a list of inputs, billed per record
`get_scraper_progress`	Get the status of a scraper run
`get_scraper_results`	Get the records of a finished scraper run
`refresh_scrapers`	Reload the scraper list now instead of waiting for the daily refresh
`list_dataset_fields`	List the filterable fields of a searchable dataset
`search_dataset`	Search a Bright Data dataset by filter and return matching records
`scraping_browser_navigate`	Open or reuse a scraping-browser session and navigate to a URL
`scraping_browser_go_back`	Go back to the previous page in the scraping browser
`scraping_browser_go_forward`	Go forward to the next page in the scraping browser
`scraping_browser_snapshot`	Capture an ARIA snapshot of the page's interactive elements
`scraping_browser_click_ref`	Click an element by its ref from the latest ARIA snapshot
`scraping_browser_type_ref`	Type into an element by ref, optionally pressing Enter
`scraping_browser_fill_form`	Fill several form fields by ARIA ref in one call
`scraping_browser_screenshot`	Capture a screenshot of the current page
`scraping_browser_network_requests`	List network requests recorded since page load
`scraping_browser_wait_for_ref`	Wait until an element identified by ref becomes visible
`scraping_browser_get_text`	Return the text content of the current page
`scraping_browser_get_html`	Return the HTML content of the current page
`scraping_browser_scroll`	Scroll to the bottom of the current page
`scraping_browser_scroll_to_ref`	Scroll until the referenced element is in view
`chatgpt_ai_insights`	Send a prompt to ChatGPT and get the answer as Markdown
`perplexity_ai_insights`	Send a prompt to Perplexity and get the answer as Markdown

Suggested access for each AI app

The documentation this list comes from doesn't mark Bright Data's tools as read-only or destructive. Read each tool's description and decide per app before allowing it.

Coding agent in Claude Code or Cursor. Allow the read tools so the agent can look things up without stopping, and require approval for tools that change shared data.

Chat or research app in Claude Desktop. Allow reading and searching. Allow writes only where you want the app to act for you.

Unattended agent, such as a scheduled Codex job. No one is watching the session, so keep it narrow: allow what the job needs, require approval for changes and block the rest.

Blocked tools are removed from the tool list an app sees, so it won't try them. Because rules apply per app, the same Bright Data connection can be broad in one app and read-only in another, and the credential stays in Toolstile instead of each app's config file.

How it works with Toolstile

Connect once. Add Bright Data to a workspace (browser sign-in). Toolstile keeps the credential, so no agent config holds it.
Choose access per app. Give Claude Code every Bright Data tool, give Cursor only the read tools, or keep Bright Data away from an app entirely.
Share approved accounts. Teammates use the connection without copying tokens, and you can remove access in one place.
Review activity. See which app called which tool and when, so you can tighten access with evidence.

Add the Bright Data MCP server to an AI app

Through Toolstile, every AI app uses one gateway entry, whichever servers you connect. In Claude Code:

Shell

claude mcp add --transport http toolstile 'https://toolstile.com/mcp'

Then connect Bright Data in your workspace and choose which of its tools that app may use. Client setup covers Codex, Cursor and Claude Desktop.

To connect directly instead, add the Bright Data endpoint to each app separately. In Claude Code:

Shell

claude mcp add --transport http brightdata 'https://mcp.brightdata.com/mcp'

In Cursor's mcp.json:

JSON

{
  "mcpServers": {
    "brightdata": {
      "url": "https://mcp.brightdata.com/mcp"
    }
  }
}

Some servers accept only clients they have approved, or need a token header. Check the Bright Data MCP documentation for what your client needs.

Before you connect

A Bright Data account. Requests consume your Bright Data usage.

Server details

Detail	Value
Publisher	Bright Data
Endpoint	`https://mcp.brightdata.com/mcp`
Authentication	Browser sign-in
Categories	Marketing & research, Data & infrastructure
Tool list	32 tools, observed 29 October 2026 (docs)

Bright Data MCP documentation

Bright Data MCP server for Claude, Codex and Cursor