n8n-nodes-mrscraper

Use MrScraper in your n8n workflows to create and run AI or manual scrapers, extract structured content, discover URLs, search the web, and retrieve results.

This package is an n8n community node for the AI-powered MrScraper web scraping platform.

Version 4.0.0

Version 4.0.0 reorganizes the node around six task-based resources and removes duplicated operations.

• New Discovery, Extraction, Results, Scraper Creation, and Scraper Runs resource names
• One Scraper Runs interface for single and batch runs of AI and manual scrapers
• Dedicated creation flows for prompt-based, listing, and website crawl scrapers
• Expected JSON output schemas for prompt-based and listing extraction
• Web Search inputs: query, country, and number of results
• Streamlined rendered-page controls with opt-in Markdown and screenshots, selectable screenshot modes, selector waits, home-page navigation, and grouped advanced browser options
• Rendered HTML requests now consistently use the MrScraper API's POST method
• More run options for General, Listing, Map, and Manual scrapers

Version 4.0.0 changes resource and operation selections. Review the migration guide before activating existing workflows.

Table of contents

• Installation
• Credentials
• Resources and operations
  • Account
  • Discovery
  • Extraction
  • Results
  • Scraper Creation
  • Scraper Runs
• Usage examples
• Migrating from 2.x
• Compatibility
• Compliance and legal risk
• Resources

Installation

Install n8n-nodes-mrscraper from Settings > Community Nodes in n8n. For other installation methods, follow the n8n community node installation guide.

Credentials

The node requires a MrScraper API credential.

1. Sign in to the MrScraper app.
2. Create or copy an API token from your account.
3. In n8n, create a MrScraper API credential and enter the token in API Token.
4. Select the credential in your MrScraper node.

Keep the token private. Do not place it directly in workflow fields or commit it to source control.

Resources and operations

Version 4.0.0 exposes six resources. Choose a resource, then select one of its operations.

Account

• Get Account Info — Get account details, token usage, and token limits.

Discovery

• Crawl Website URLs — Discover links from a starting URL. Set crawl depth, pages evaluated, result limit, and optional pipe-separated include or exclude regular expressions.
• Web Search — Search the web using a query, country (default us), and number of results (default 10).

WebSearch is a tool that searches the web to retrieve real-time results, including relevant website URLs, titles, descriptions, and related questions based on a given query.

Extraction

• Extract Page by Prompt — Extract data from one page using a natural-language prompt and an optional expected JSON output schema. Supports Super or Cheap mode and an optional proxy country.
• Extract Listings and Paginated Content — Extract repeated listing data with a prompt, optional item schema, page limit, and proxy country.
• Extract Structured Data — Use a preset extraction schema for Article, Forum Thread, Hotel, Job Posting, Post, Product, Property, Restaurant, Social Media Profile, or Tour / Attraction content.
• Fetch Rendered HTML — Load a page using the MrScraper stealth browser with a rendered JavaScript page. HTML is returned by default; Markdown and screenshots are opt-in. A screenshot can cover the full page or only the top. The primary fields configure retries, timeout, geolocation, and proxy country, while Advanced Options contains the token cap, CSS selector wait, lifecycle wait condition, resource blocking, home-page navigation, cookies, and Super mode.

Results

• Get Results — Get paginated results for a scraper ID with page size and created-date sort order.
• Get Latest Results — Get the latest N results for a scraper ID.
• Get Result Detail — Get one complete result by result ID.

Scraper Creation

• Create Prompt-Based Scraper — Create a General AI scraper from a URL, prompt, optional expected JSON schema, mode, and proxy country.
• Create Listing Scraper — Create a Listing AI scraper from a URL, prompt, optional item schema, maximum page count, and proxy country.
• Create Website Crawl Scraper — Create a Map AI scraper with crawl depth, maximum pages, result limit, and optional URL include or exclude patterns.

Created scrapers are available in your MrScraper account and can be reused with Scraper Runs.

Scraper Runs

• Run Existing Scraper — Run one URL with an existing AI or manual scraper.
• Run Existing Scraper in Batch — Run multiple URLs with an existing AI or manual scraper.

For a single AI run, choose the agent type:

• General — Prompt-based page extraction, with optional HTML, Markdown, screenshot, cookies, JavaScript rendering, selector wait, and home-page visit settings.
• Listing — Listing and pagination extraction, with page and timeout limits plus optional streaming and rendering settings.
• Map — Website URL discovery with crawl limits and URL filtering patterns.

Manual runs support browser, proxy, cookie, paginator, recording, output, timeout, and token-cap options. Batch runs accept URLs as a JSON array or a comma/newline-separated list.

Usage examples

Search the web

1. Select Discovery > Web Search.
2. Select your saved MrScraper API credential.
3. Enter Search Query, for example best hotels in New York.
4. Set Country (default us) and Number of Results (default 10), then run the node.

The node sends POST https://api.mrscraper.com/ with action: "websearch", query, country, and numResults. Authentication uses the saved credential in the x-api-token header.

Extract data with a JSON shape

1. Select Extraction > Extract Page by Prompt.
2. Enter the page URL.
3. Enter a prompt such as Extract the product name, price, availability, and image URL.
4. Optionally enter an expected output schema:

{
    "name": "string",
    "price": "number",
    "inStock": "boolean",
    "imageUrl": "string"
}

5. Select Super for stronger scraping capabilities or Cheap for sites with weaker protection, then run the node.

Fetch a rendered page

1. Select Extraction > Fetch Rendered HTML and enter the target URL.
2. Leave Return HTML enabled, or enable Return Markdown if Markdown output is also needed.
3. To capture an image, enable Screenshot, then choose Full or Top under Screenshot Mode.
4. Use Advanced Options when the page needs a CSS selector wait, a different lifecycle event, resource blocking controls, home-page navigation, returned cookies, Super mode, or a custom token cap.

The operation defaults to a 300-second timeout, three retries, us geolocation and proxy country, HTML output enabled, and Markdown and screenshot output disabled. Wait Until supports DOM Content Loaded, Load, and Network Idle.

Run an existing scraper in batch

1. Select Scraper Runs > Run Existing Scraper in Batch.
2. Choose AI or Manual and enter the scraper ID.
3. Enter the URLs as a JSON array:

["https://example.com/page-1", "https://example.com/page-2"]

You can also use comma-separated or newline-separated URLs.

Retrieve completed results

Use Results > Get Latest Results with the scraper ID, or use Get Result Detail when you already have a result ID.

Migrating from 2.x

After upgrading, open each existing MrScraper node and reselect its resource and operation. The most common mappings are:

Also review these field changes:

• Web Search replaces the legacy search operation (syncSerp). Reselect Discovery > Web Search in existing search nodes and configure query, country, and numResults; region, language, page, response format, and JavaScript rendering are no longer search inputs.
• Single and batch reruns now share Scraper Runs. Select AI or Manual explicitly.
• AI single runs expose agent-specific fields after selecting General, Listing, or Map.
• Batch URLs can be a JSON array string or a comma/newline-separated list.
• Prompt-based and listing operations can append an expected JSON schema to the extraction prompt.
• Fetch Rendered HTML now uses a screenshot toggle followed by a Full or Top mode; screenshots and Markdown are disabled by default.
• Browser rendering is always enabled and is no longer a user-configurable field.
• Token cap, resource blocking, lifecycle waiting, cookies, and Super mode are now under Advanced Options, which also adds CSS selector waiting and optional home-page navigation.

Test migrated workflows before enabling them in production.

Compatibility

• n8n-nodes-mrscraper version: 4.0.0
• Node.js: 22 or newer
• Package manager used by this project: npm

Compliance and legal risk

[!WARNING] Scraping login-protected pages carries serious legal and compliance risks. Many websites prohibit automated access in their terms of service. Only scrape content you are authorized to access, review the target website's terms and applicable laws, and accept responsibility for how you use this node.

Resources

• MrScraper
• MrScraper app
• MrScraper API documentation
• MrScraper programmatic guide
• MrScraper n8n integration
• n8n community nodes documentation
• Changelog

License

This package is released under the MIT License.