Log inSubmit server
Crawlbrulee MCP logo

Crawlbrulee MCP

by crawlbrulee

EU-based web scraping over MCP with page extraction, site mapping, async jobs, usage checks, and account identification.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Crawlbrulee MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "crawlbrulee-crawlbrulee-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@crawlbrulee/mcp"
      ],
      "env": {
        "CRAWLBRULEE_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Install not yet sandbox-tested— the server itself is confirmed live (we called its tools/list endpoint). The stdio install command is a separate, ongoing check.

Overview

crawlbrulee/crawlbrulee-mcp connects MCP-compatible agents to the Crawlbrulee web scraping API through a stdio server. It provides single-page extraction, CSS-selector-based value extraction, site-wide URL mapping, asynchronous jobs, usage reporting, and API-account inspection. Reach for it when an agent needs structured web content, screenshots, links, metadata, or controlled crawling rather than only browser interaction. Results include usage information and status details that help agents handle billing, incomplete output, HTTP error pages, and unreachable targets.

Use cases

  • Extract structured fields from product or article pages
  • Map a website before selecting pages to scrape
  • Run long JavaScript-heavy scrapes asynchronously
  • Collect page screenshots, metadata, links, or cleaned HTML
  • Check credits and the active Crawlbrulee account before crawling

Key features

  • Single-page extraction with multiple output formats
  • CSS-selector-based named value extraction
  • Paginated site URL discovery
  • Asynchronous jobs with status polling and webhooks
  • Usage and authenticated-account inspection
  • Cleanup, proxy, location, caching, and zero-retention controls

Tools7

What Crawlbrulee MCP lets your agent do. Expand a tool for its parameters.

Verified live~8.1k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

10/10 checks healthy over the last 44d

What it can do

From the 7 tools this server returned when we ran it.

Handles credentials · 2Makes network requests · 6Read-only tools · 7
Network at startup3 connections
Files written at startupNone
Install scriptsNone
npm provenanceSigned build
Package age150 days
Maintainers1

Observed 2026-10-27 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 14d ago via OSV.dev · @crawlbrulee/mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads1.1kPackage downloads in the last 30 days.
Last commit5d agoMost recent push to the default branch.
Tools exposed7Callable tools this server registers over MCP.
Directory activity2 viewsConfig copies, upvotes, and views on AllMCPs.

How Crawlbrulee MCP works

What crawlbrulee/crawlbrulee-mcp does

crawlbrulee/crawlbrulee-mcp exposes Crawlbrulee's web scraping functions as MCP tools for agents and developer workflows. A synchronous scrape can return markdown, cleaned HTML, raw HTML, links, images, screenshots, and page metadata. The server also supports named extraction: provide a CSS selector for values such as headings, prices, or links, and receive them under matching names without an additional credit charge.

For broader discovery, map combines sitemap reading with homepage-link extraction and returns normalized URLs in pages. The result can be filtered for internal, external, or subdomain links. usage reports the authenticated organization's credit balance and concurrency information, while whoami identifies the organization and token associated with the configured API key.

How it works

The server runs over stdio and uses the Crawlbrulee SDK behind the MCP adapter. scrape is intended for immediate fetches. Longer operations, including heavy JavaScript rendering or large full-page screenshots, can be submitted with scrape_async; poll with scrape_status, then retrieve the completed payload with scrape_result. A completion webhook can replace polling for asynchronous jobs.

Scrape responses expose the final page's HTTP status separately from tool-level failures. A served 404 or 503 is returned as content with its status, whereas a target that cannot be reached produces a target_unreachable error. Responses also include the resolved engine, proxy tier, and credit calculation. The service can clean ads, cookie banners, chat widgets, and selected CSS regions before building most outputs; raw HTML remains unmodified by that cleanup.

Setup and configuration

Create a Crawlbrulee API token, place it in the environment, and launch the package with an MCP host:

server.ts
export CRAWLBRULEE_API_KEY="cwbl_..."
npx -y @crawlbrulee/mcp

The sole required environment variable is CRAWLBRULEE_API_KEY. The server reads it when the first tool is called, so authentication errors appear as tool errors rather than preventing startup. Claude Code can register the command directly with claude mcp add crawlbrulee -- npx -y @crawlbrulee/mcp; other stdio hosts use the same executable and may place the variable in their configuration.

Tools and capabilities

  • scrape: Fetch one URL and select output types, rendering, proxy, location, caching, cleanup, screenshots, or named CSS-selector extraction.
  • scrape_async: Start a long-running scrape and optionally register a signed completion webhook.
  • scrape_status: Check whether an asynchronous job is pending, running, complete, or failed.
  • scrape_result: Retrieve the output of a completed asynchronous job.
  • map: Discover and paginate site URLs with discovery budgets and link-type filters.
  • usage: Inspect cycle credits, quota percentage, concurrency, and reset timing.
  • whoami: Confirm the organization, token name, and shortened token preview for the configured key.

Limitations and notes

Large responses can be truncated: limits include 30,000 links, 10,000 inline images, and 10,000,000 HTML characters. Check warnings for truncation or unavailable sections before treating output as complete. Screenshot links are signed and expire after 24 hours. Asynchronous jobs and their results are also available for 24 hours.

Zero-data-retention mode excludes results from the shared cache, keeps delivery data for 24 hours, and adds one credit; the organization must enable it first. Map discovery and scrape operations report credit usage, and page billing depends on the returned status and selected engine. Use page_status_code to distinguish requested content from a site's error page.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Crawlbrulee MCP

crawlbrulee/crawlbrulee-mcp is an MCP server that connects AI agents to the Crawlbrulee web scraping API. Its main tools scrape individual pages, map websites, run asynchronous scrape jobs, check job status and results, report organization usage, and identify the configured API account.

Install it by setting `CRAWLBRULEE_API_KEY` and running `npx -y @crawlbrulee/mcp` through an MCP host. Claude Code can register that command with `claude mcp add crawlbrulee -- npx -y @crawlbrulee/mcp`; other stdio hosts use the same command.

Yes, Cursor is named as a supported MCP host for this server. Add the stdio command `npx -y @crawlbrulee/mcp` to Cursor's MCP configuration and provide the API key or its environment variable there.

Use `scrape_async`, then call `scrape_status` until the job is `done` and retrieve it with `scrape_result`. This avoids holding the connection open; a completion webhook can be registered if polling is not preferred.

Related MCP servers

All in Search & Data Extraction All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewCrawlbrulee MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/crawlbrulee-crawlbrulee-mcp?style=directory)](https://allmcps.com/mcp/crawlbrulee-crawlbrulee-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/crawlbrulee-crawlbrulee-mcp"><img src="https://allmcps.com/api/badge/crawlbrulee-crawlbrulee-mcp?style=directory" alt="Crawlbrulee MCP on AllMCPs" /></a>