Integration · Research and analyze

Bright Data and Meteor AI agents

Connect Bright Data to a Met: search Google, read protected pages, extract data as JSON and run ready made scrapers with the official MCP.

Book a demo See every integration

How does Bright Data connect to Meteor?

With the Bright Data ready connector, your Met uses Bright Data's infrastructure on your account, authorized once with OAuth. It searches Google, Bing or Yandex, reads a web page as clean text even if it has bot protection, and turns its content into structured data. It also searches Bright Data's ready made scrapers, for example for reviews or for a store's products, runs them and returns the results, and drives a browser for pages that need clicking or typing; running scrapers, clicking and typing start off until the admin turns them on. Every new account gets free requests each month and ready made scrapers are billed per record on your Bright Data account.

What it automates

What a Meteor Met does with Bright Data

Check competitor prices

You ask the Met how much the competition charges for a product. It finds the pages, reads them even if they have bot protection and sums up the prices it found with their links.

Bring in a product's reviews

The Met finds the ready made reviews scraper for the store you care about and, if the admin turned on running it, runs it on the links you give and tells you what keeps coming up in the reviews.

Turn a page into data

Instead of copying a table from a website by hand, the Met reads the page and turns it into structured data you can save or compare.

Browse a page that needs interaction

If the data only shows up after clicking or typing a search, the Met opens a browser, takes those steps and reads the text that remains. Clicking and typing start off.

20 tools

Everything your Met can ask Bright Data for

Each one is a real action in the connector. You can switch off the ones you would rather it never used.

Tool	What it does
`search_engine`	Returns search results from Google, Bing or Yandex, with pagination.
`search_engine_batch`	Runs up to 10 searches at once.
`discover`	Searches the web and ranks results by how well they match what is being looked for.
`scrape_as_markdown`	Reads a web page and returns it as clean text, even if it has bot protection.
`scrape_batch`	Reads up to 10 pages in a single request.
`scrape_as_html`	Reads a protected web page and returns its HTML.
`extract`	Reads a page and turns its content into structured data, with optional instructions.
`search_scrapers`	Finds a ready made scraper by site or data type, for example a store's reviews.
`get_scraper_details`	Shows what input a ready made scraper needs and which fields it returns.
`run_scraper`	Runs a ready made scraper on a list of inputs. Billed per record. It starts off until the admin turns it on.
`get_scraper_progress`	Tells whether a scraper run is in progress, ready, failed or cancelled.
`get_scraper_results`	Returns the records of a finished scraper run.
`scraping_browser_navigate`	Opens a browser session and goes to the given address.
`scraping_browser_snapshot`	Takes a snapshot of the page structure with the elements that can be acted on.
`scraping_browser_click_ref`	Clicks an element of the page identified in the latest structure snapshot. It starts off until the admin turns it on.
`scraping_browser_type_ref`	Types into a field on the page and, if asked, presses Enter. It starts off until the admin turns it on.
`scraping_browser_get_text`	Returns the text of the page open in the browser.
`scraping_browser_screenshot`	Takes a screenshot of the page, full length if asked.
`scraping_browser_scroll`	Scrolls to the bottom of the page open in the browser.
`scraping_browser_go_back`	Goes back to the previous page in the browser session.

The limits

What the Bright Data integration does NOT do

We would rather say it here than have you find out during rollout.

It is not free without limit. Every new account gets free requests each month; beyond that, and for ready made scrapers billed per record, the cost runs on your Bright Data account.
It does not decide whether you may use a piece of data. Reading a public page does not give you permission to reuse its content; you check that against each site's terms.
It does not ask you for confirmation in the chat before acting. The admin chooses which actions stay on; running scrapers, clicking and typing in the browser start turned off.
It does not wait forever for a scraper. The run waits up to 45 seconds; if it is still going, it returns an identifier and the Met fetches the results later.

Before you start

What you need to connect Bright Data

1

A Bright Data account

You create your account in Bright Data. Every new account gets free requests each month; ready made scrapers are billed per record.
2

Connect it in Met

In Met, open Connections → Bright Data → Connect. You sign in to Bright Data and authorize access; there is no token to paste.
3

Decide which actions stay on

The workspace admin chooses which actions stay on. Running scrapers, which are billed per record, and clicking or typing in the browser start turned off.

Frequently asked questions

Questions about Bright Data and Meteor

Bright Data bills it on your account. Its documentation says every new account gets five thousand free requests per month and that ready made scrapers are billed per record.

Yes. Reads go through Bright Data's infrastructure, which handles connection rotation and the page's anti bot challenges.

It is a scraper Bright Data has already built for a site or a data type. The Met finds it, checks which inputs it needs, runs it and returns the records.

You do, with your Bright Data account. Met does not charge for the calls your Met makes to Bright Data.

Step by step