Appearance
n8n, Make and Zapier
Use ScrapeIt from the automation tool you already work in. The n8n node, the Make app and the Zapier app call the same REST API you would call yourself: the same proxies, anti-bot handling, concurrency limits and billing outcome rules. For AI assistants and agents (Claude, Cursor, VS Code), connect the MCP server instead.
| Platform | What you get | Install |
|---|---|---|
| n8n | Community node n8n-nodes-scrapeit: Web Scraping, Google Search, Google Maps (search, place), Amazon (search, product), Crawler (start, status), Job (get). | Settings → Community nodes → Install, package n8n-nodes-scrapeit. |
| Make | ScrapeIt app: Scrape a URL, Search Google, Search places on Google Maps, Get an Amazon product, Start a website crawl, Get a job. | Add a module in the scenario editor and search for ScrapeIt. |
| Zapier | ScrapeIt app: actions Scrape URL, Google Search, Get Amazon Product, Start Crawl; trigger Job Finished. | Add a step in the Zap editor and search for ScrapeIt. |
| MCP | Every ScrapeIt tool for AI assistants and agents. | Point the client at https://api.scrapeit.io/api/v1/mcp - see MCP server. |
Create an API key for the integration
All three integrations sign in with an API key. Create one in Workspace → API Keys, a separate key for each tool, so a leaked key has a small blast radius and usage stays attributable (security practices). Give it the scopes of the steps you use:
| Scope | Needed for |
|---|---|
scrape:web | Scrape URL, Google Search, Google Maps, Amazon. |
scrapers:run | Start Crawl. |
scrapers:read | Crawl status, Get a job and the Zapier Job Finished trigger. |
usage:read | The connection test each tool runs when you add the key - it spends no credits. |
The key is stored by n8n, Make or Zapier in their encrypted credential store and is sent only in the x-api-key header to https://api.scrapeit.io.
n8n
- In n8n, open Settings → Community nodes → Install, enter
n8n-nodes-scrapeitand confirm. On self-hosted n8n this needs an owner or admin account. - Add the ScrapeIt node to a workflow, open Credential → Create new and paste your API key. Leave Base URL at
https://api.scrapeit.io. - Pick a resource and an operation. Optional API parameters are under Options.
| Resource | Operation | API endpoint |
|---|---|---|
| Web Scraping | Scrape URL | POST /api/v1/scrape/web |
| Google Search | Search | GET /api/v1/scrape/google/search |
| Google Maps | Search Places | GET /api/v1/scrape/google/maps/search |
| Google Maps | Get Place | GET /api/v1/scrape/google/maps/place |
| Amazon | Search | GET /api/v1/scrape/amazon/search |
| Amazon | Get Product | GET /api/v1/scrape/amazon/product |
| Crawler | Start Crawl | POST /api/v1/scrapers/crawler/jobs |
| Crawler | Get Status | GET /api/v1/scrapers/jobs/{jobId} |
| Job | Get | GET /api/v1/jobs/{jobId} |
Field names match the API: JS Rendering is auto, on or off (JavaScript rendering), Extract Rules takes the same JSON as extractRules, and crawler Seed URLs and path patterns accept a comma- or newline-separated list. Each output item is the response's data object. The node can also be used as a tool by n8n AI agents.
Example workflow - daily rank check: Schedule Trigger → ScrapeIt (Google Search, Query best coffee maker, Options → Country de) → Split Out on organic_results → Google Sheets Append row with position, title and link.
Make
- In a scenario, add a module, search for ScrapeIt and pick an action.
- Click Create a connection and paste your API key. Make checks it right away.
- Fill in the fields; empty optional fields are not sent, so the API defaults apply.
| Module | API endpoint |
|---|---|
| Scrape a URL | POST /api/v1/scrape/web |
| Search Google | GET /api/v1/scrape/google/search |
| Search places on Google Maps | GET /api/v1/scrape/google/maps/search |
| Get an Amazon product | GET /api/v1/scrape/amazon/product |
| Start a website crawl | POST /api/v1/scrapers/crawler/jobs |
| Get a job | GET /api/v1/jobs/{jobId} |
Example scenario - price tracker: Google Sheets Watch Rows (one ASIN per row) → ScrapeIt Get an Amazon product → Google Sheets Update a Row with price, availability and rating.
Example scenario - crawl and collect: ScrapeIt Start a website crawl → Tools Sleep → ScrapeIt Get a job with the crawl id, repeated until status is success, failed or stopped → download the json_results file through the Files API.
Zapier
- In the Zap editor, add a step, search for ScrapeIt and pick an action or the trigger.
- Connect your account with the API key.
| Step | API endpoint |
|---|---|
| Scrape URL (action) | POST /api/v1/scrape/web |
| Google Search (action) | GET /api/v1/scrape/google/search |
| Get Amazon Product (action) | GET /api/v1/scrape/amazon/product |
| Start Crawl (action) | POST /api/v1/scrapers/crawler/jobs |
| Job Finished (polling trigger) | GET /api/v1/jobs |
Job Finished checks the newest jobs of the type you choose (crawls, batch jobs, no-code jobs or single API requests) on your Zapier plan's polling interval and fires once per job that reached success, failed, blocked, timeout or stopped.
Example Zap - crawl report to Slack: ScrapeIt Job Finished (Job Type: Website crawls) → Slack Send Channel Message with the job status, credits and targetUrl.
Zapier stops a step after 30 seconds
Zapier gives every action 30 seconds. Scrape URL therefore sends timeoutMs: 25000: a page that takes longer ends as SCRAPE_TIMEOUT instead of running on after Zapier gave up. For slow, rendered or residential targets use n8n or Make, which wait up to the API's full 180-second budget, or start a crawl and continue on Job Finished.
Instant events with webhooks
Polling waits for the next check. To react the moment a crawl ends, add a webhook endpoint in the dashboard that points at an n8n Webhook node, a Make Custom webhook or a Zapier Catch Hook URL and subscribe it to scraper.job.finished and scraper.job.failed. Verify the signature as described in Webhooks before trusting the payload.
Credits, limits and errors
- Every step run is one API request and costs exactly what a direct call costs (billing & credits).
- Your plan's concurrency limit counts integration requests too. A step that hits it fails with
CONCURRENCY_LIMIT_EXCEEDED(HTTP 429) before any work starts. Zapier retries it later on its own; in n8n turn on Retry On Fail, in Make add a Break error handler with retries. - Errors show the stable code and message, for example
INSUFFICIENT_CREDITS: …- see error codes. A revoked or wrong key (INVALID_API_KEY) asks you to reconnect the account.
Related pages
- MCP server - ScrapeIt tools for Claude, Cursor, VS Code and other AI clients.
- Authentication & API keys - scopes, rotation and key hygiene.
- Web Scraping API - every parameter of Scrape URL.
- Website Crawler API - crawl parameters and result files.
- Getting job results - polling vs webhooks.