MCP For Claude Marketplace
Bright Data

Bright Data MCP Connector for Claude

A+

Access the world's #1 web data platform — bypass anti-bot protections, extract structured search engine data, and manage scraping browsers directly from your AI agent.

10 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

Connect your Bright Data account to any AI agent to automate complex web data extraction, bypass sophisticated anti-bot systems, and manage your proxy infrastructure through natural language.

What you can do

  • Web Unlocking & SERP — Use the send_request tool to bypass anti-bot protections or extract structured data from search engines using Web Unlocker or SERP API zones.
  • Dataset Orchestration — Trigger large-scale data collections with trigger_dataset, monitor progress with get_dataset_progress, and download snapshots once ready.
  • Browser Management — List and monitor active Scraping Browser sessions to debug or track automated browser workflows.
  • Infrastructure Control — Retrieve all configured zones, check account status, and fetch proxy credentials (passwords) for specific zones securely.
  • Marketplace Discovery — Browse available datasets in the Bright Data marketplace to find pre-collected data for your industry.

How it works

  1. Subscribe to this server
  2. Enter your Bright Data API Key
  3. Start scraping and managing proxies from Claude, Cursor, or any MCP-compatible client

Your AI now acts as a data engineer, capable of retrieving real-time web information even from the most protected sources.

Who is this for?

  • Data Scientists & Analysts — Automate the collection of training data or market intelligence without manual script execution.
  • Developers — Debug scraping browser sessions and manage proxy zones directly from the code editor.
  • Market Researchers — Extract structured SERP data to track rankings and competitor movements via simple conversation.
proxy-managementserp-apiweb-unlockerdata-extractionbrowser-automation

10 tools expose this connector's capabilities to your AI agent.

get_dataset_snapshot

Returns structured JSON records. Error records include error_code: "dead_page" (404) or "bad_input" (wrong URL pattern). Download scraped data from a completed collection

list_datasets

Use to find the correct dataset_id for trigger_dataset. Large response — filter by name. Browse all 100+ available datasets with IDs and names

get_zone_info

Use to verify a zone is active and correctly configured before using send_request. Get zone configuration — type, status, bandwidth, and settings

get_all_zones

Call this before send_request to get valid zone names. Empty list means zones must be created at brightdata.com/cp/zones. List all proxy zones configured in the account

get_dataset_progress

Returns status ("running" or "ready"), records count, and errors count. Poll every 15-30s. Only call get_dataset_snapshot after status="ready". LinkedIn scraping takes 60-120s per URL. Check if a dataset collection is complete

get_zone_passwords

For direct proxy connections outside this MCP. SENSITIVE — do not log credentials. Get proxy credentials for direct connections (Selenium, Playwright, etc.)

get_account_status

Use first if other tools return auth errors. Check account status and API key validity

list_browser_sessions

Only relevant if Scraping Browser zones are configured. Default: 50 most recent. List Scraping Browser sessions by status, duration, or bandwidth

send_request

Requires a valid zone name — call get_all_zones first if unknown. Web Unlocker zones scrape any site; SERP API zones return structured search results from Google/Bing URLs. Scrape any URL bypassing anti-bot protections, or extract structured SERP data

trigger_dataset

Poll get_dataset_progress until status="ready", then call get_dataset_snapshot to download results. Provide either url or keyword. LinkedIn Posts (gd_lyy3tktm25m4avu764) requires URLs matching linkedin.com/(pulse|posts|feed/update) — no keyword support. Always set include_errors=true. Start an async scraping job for LinkedIn, Amazon, Instagram, and 100+ sources

See how to talk to your AI agent using Bright Data.

Scrape this LinkedIn post and extract the engagement data.

I triggered the LinkedIn Posts dataset with your URL and received snapshot `sd_abc123`. After polling for 90 seconds, the collection completed with 1 record and 0 errors. The post has 142 reactions, 23 comments, and was published 3 days ago. Here is the full structured data.

What proxy zones do I have available, and scrape this protected page.

You have 2 active zones: `web_unlocker1` (Web Unlocker) and `serp_api1` (SERP API). I used `web_unlocker1` to scrape the protected page and retrieved the full HTML content converted to markdown. The page contains 3 product listings with prices and reviews.

Find all available LinkedIn-related datasets I can use.

I found 6 LinkedIn datasets: LinkedIn Posts (`gd_lyy3tktm25m4avu764`), LinkedIn People Profiles (`gd_l1viktl72bvl7bjuj0` — 115M profiles), LinkedIn Companies (`gd_l1vikfnt1wgvvqz95w` — 55M companies), LinkedIn People Enriched (`gd_me5ppxjr2ge6icjuh0`), LinkedIn Companies with Jobs (`gd_mhf4aje023ctc3yn7w`), and LinkedIn Profiles by Position (`gd_la9vypp0jzg1bhkv3`). Which one would you like to use?

Use `trigger_dataset` with the LinkedIn Posts dataset ID (`gd_lyy3tktm25m4avu764`) and a direct post URL (e.g., `https://www.linkedin.com/feed/update/urn:li:activity:...`). This returns a `snapshot_id`. Then poll `get_dataset_progress` every 15–30 seconds until the status is `ready` — LinkedIn scraping typically takes 60–120 seconds. Finally, call `get_dataset_snapshot` to retrieve the structured data including post content, author details, reactions, and comment count.

Related Connectors

Olostep

Olostep MCP

10 tools Official

Scrape web pages at scale with a headless browser API that renders JavaScript and returns clean structured data instantly.

A+ View details →
Browse AI

Browse AI MCP

12 tools Official

Extract data from any website without code using trained robots that monitor pages and deliver structured results automatically.

A+ View details →
Decodo (Smartproxy)

Decodo (Smartproxy) MCP

9 tools Official

Manage your Decodo (formerly Smartproxy) network infrastructure — list proxy endpoints, monitor bandwidth, and fetch residential proxy configurations with your AI.

A+ View details →
Nimbleway

Nimbleway MCP

10 tools Official

Web data collection and scraping via Nimbleway — extract content and search the web directly from your AI agent.

A+ View details →