# brightdata-mcp [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/brightdata/brightdata-mcp  
**GitHub Stars:** 2686  
**npm Downloads (last month):** 46951  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/brightdata-mcp

## Description
Bright Data's Web MCP server enabling AI agents to search, extract & navigate the web

## Tools
Capabilities this server exposes over MCP:

- **search_engine** — Scrape search results from Google, Bing or Yandex. Returns SERP results in JSON or Markdown (URL, title, description),Ideal forgathering current information, news, and detailed search results.
- **scrape_as_markdown** — Scrape a single webpage URL with advanced options for content extraction and get back the results in MarkDown language. This tool can unlock any webpage even if it uses bot detection or CAPTCHA.
- **search_engine_batch** — Run multiple search queries simultaneously. Returns JSON for Google, Markdown for Bing/Yandex.
- **scrape_batch** — Scrape multiple webpages URLs with advanced options for content extraction and get back the results in MarkDown language. This tool can unlock any webpage even if it uses bot detection or CAPTCHA.
- **discover** — Search the web and rank results by AI-driven relevance. Returns scored results with title, description, and URL. Supports intent-based ranking, geo-targeting, date filtering, and keyword filtering.
- **search_scrapers** — Find a Bright Data pre-built scraper (Web Scraper API, 1,200+ sites) for a site or data type, e.g. "amazon reviews" or "linkedin.com". Returns dataset_id, name, domain and collection_methods. Next: call get_scraper_details with a dataset_id and one method.
- **get_scraper_details** — Show the scraper description, input fields, example input and main output fields for one method. Call before run_scraper and build the input from sample_input or input_fields. Records may contain more fields than output_fields lists.
- **run_scraper** — Run a scraper method. input is a list of objects shaped like sample_input from get_scraper_details, one per URL or query. collect_by_url returns one record per input; discover_by_* methods return up to limit_per_input records per input, and every record is billed. Waits up to 45 seconds; if still running it returns snapshot_id with status "running": then poll get_scraper_progress and call get_scraper_results when ready. Records with an "error" field explain failed inputs.
- **get_scraper_progress** — Status of a run: starting, running, ready, failed or canceled.
- **get_scraper_results** — Records of a finished run. Returns status "running" if not ready yet.
- **refresh_scrapers** — Reload the scraper list now instead of waiting for the daily refresh.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "brightdata-mcp": {
    "command": "npx",
    "args": ["@brightdata/mcp"],
    "env": {
      "API_TOKEN": ""
    }
  }
}
```

**Requires environment variables:** `API_TOKEN` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What brightdata-mcp MCP server does

The brightdata-mcp MCP server gives MCP-compatible AI clients access to Bright Data's web search and extraction services. Its tools cover three main workflows: finding current information, converting webpages into Markdown, and collecting structured records through Bright Data's pre-built scraper catalog.

`search_engine` queries Google, Bing, or Yandex and returns result fields such as URLs, titles, and descriptions in JSON or Markdown. `search_engine_batch` runs several queries together. `discover` adds AI-based relevance ranking, intent-based ranking, geographic targeting, date filters, and keyword filters to web discovery.

For page content, `scrape_as_markdown` handles one URL and `scrape_batch` handles multiple URLs. Both support advanced extraction options and return Markdown.

## How it works

You can connect through Bright Data's hosted MCP endpoint or run the package locally with `npx @brightdata/mcp`. The hosted option uses a URL containing an API token; the local option receives the token through the `API_TOKEN` environment variable.

Structured extraction follows a separate sequence. First, `search_scrapers` finds a suitable pre-built scraper by site or data type. Then `get_scraper_details` returns its description, accepted fields, sample input, and primary output fields. Build the request from that information before calling `run_scraper`.

A scraper run accepts a list of input objects. It may return records immediately or provide a `snapshot_id` while the job remains in progress. Use `get_scraper_progress` to check states such as `starting`, `running`, `ready`, `failed`, or `canceled`, and use `get_scraper_results` when the run is ready. `refresh_scrapers` reloads the scraper catalog instead of waiting for its daily refresh.

## Setup and configuration

The hosted endpoint is `https://mcp.brightdata.com/mcp?token=YOUR_API_TOKEN`. The README provides setup examples for Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Gemini CLI, and Zed. Hosted connections can optionally restrict the available tools with `groups` or `tools` URL parameters.

For a local stdio connection, configure the client to run `npx @brightdata/mcp` and set `API_TOKEN` to a Bright Data account token. Scraper runs may bill each returned record, and a run that exceeds the stated wait period can remain asynchronous.

## Tools and capabilities

- Search Google, Bing, or Yandex and return SERP data.
- Run multiple search or page-extraction requests in a batch.
- Extract individual webpages or URL sets as Markdown.
- Rank web results by relevance with filters for intent, location, date, and keywords.
- Find, inspect, run, monitor, and retrieve results from Bright Data pre-built scrapers.
- Refresh the available scraper catalog on demand.

The repository describes the broader package as covering web search, page scraping, structured extraction, and browser automation, but the tools listed for this entry are the search, extraction, discovery, and scraper-management tools above.

_Full upstream README: https://allmcps.com/mcp/brightdata-mcp/readme_