Web Scraping API

This chapter covers how Web Scraping API works, when Credits are charged, parameters, concurrency, and common developer questions.

How it works

What is Web Scraping API vs a traditional proxy?

Web Scraping API is a hosted REST fetch for collection jobs.

Traditional proxy: you get a tunnel. You run Puppeteer / Playwright / Selenium, defeat JS and WAF, forge headers, and retry on timeouts.
Web Scraping API: send an HTTP request with the target URL. The service picks an exit, uses a real-looking fingerprint, handles common anti-bot, runs client JS, and returns HTML or JSON.

Why “pay only on success”? How billing works

JoyProxy charges only for successful requests:

A usable 2xx body deducts Credits.
Target blocks, timeouts, failed challenges, or network blips do not deduct Credits.
Purchased Credits do not expire.

How do I buy Credits? What packs exist?

Sign in, open Web Scraping API, go to Buy, and pay with Balance. Seven preset packs plus custom volume (10,000 Credits minimum):

Micro: $5.00 / 10,000 Credits (~$0.50 / 1K standard fetches, max concurrency 5)
Starter: $12.00 / 28,000 Credits (~$0.43 / 1K, concurrency 10)
Hobby: $39.00 / 100,000 Credits (~$0.39 / 1K, concurrency 20)
Growth: $99.00 / 300,000 Credits (~$0.33 / 1K, concurrency 40)
Pro: $249.00 / 1,000,000 Credits (~$0.25 / 1K, concurrency 75)
Business: $599.00 / 4,000,000 Credits (~$0.15 / 1K, concurrency 120)
Scale: $1,500.00 / 15,000,000 Credits (~$0.10 / 1K, concurrency 200)

How many Credits per request?

Cost depends on options:

Request type	Parameters	Credits	Notes
Standard	Default (datacenter)	1 Credit	Static sites and open APIs with light bot defense
JS rendering	`render=true`	5 Credits	Headless browser, wait for DOM
Residential / mobile exit	`super=true`	10 Credits	Real residential/mobile pool for strict IP reputation
Residential + JS	`super=true` and `render=true`	25 Credits	Tightest e-commerce, social, and travel sites

Official structured plugins?

First-party plugins skip your own parsers:

Endpoint and auth?

One hosted fetch URL:

URL: https://api.joyproxy.com/v1/fetch
Methods: GET and POST (for POST use Content-Type: application/json)
Auth: ?token=YOUR_TOKEN, or Authorization: Bearer <YOUR_TOKEN>, or X-Token: <YOUR_TOKEN>.

What is Scraping API Token? Where do I copy it?

Scraping API Token is isolated from the site password, extract API token, and AI Access Token.

After you buy Credits, open Web Scraping API → API Center to copy it. Click Rotate token if it may have leaked.

Common parameters? Sticky sessions?

On /v1/fetch:

url (required): full target URL.
geoCode (optional): ISO-2 country (us, gb, jp); 200+ countries.
render (optional): true to render JS and wait for AJAX.
super (optional): true for residential/mobile exits.
sessionId (optional): sticky context. Reuse the same string (for example sessionId="cart_step_1") for a few minutes so login + next pages or add-to-cart stay on one session.

Async queue?

Yes, for large jobs or slow renders:

POST /v1/fetch/async/jobs: submit a batch; returns job_id and task_id list;
GET /v1/fetch/async/jobs/{job_id}: poll progress;
GET /v1/fetch/async/jobs/{job_id}/{task_id}: one task result.

What is in the Web Scraping API console?

Four tabs on admin-web-unblocker.html: