Sites and Data Fields

A site is a source and a boundary for its related scenarios, runs, records, and exports. Click the plus button in Sites, then enter a recognizable name and address. The https:// protocol is selected by default; switch to http:// for sites without HTTPS. You can paste a complete URL and the form will detect its protocol. A description is optional.

The site icon is found automatically. You can upload your own icon during creation, even when an automatic one was found. PNG, JPEG, and WebP images up to 4 MB and 2048 × 2048 pixels are accepted. GIF is not accepted; the saved icon remains static. You can replace it later in the site settings. The name and description are filled from the page if you have not entered them yourself. The system generates the site code, which remains stable for links and integrations.

Collection settings

The main settings appear at the top of the site settings page. Connection, frequency, and browser options are under Additional collection settings. If you are just getting started, you do not need to change them.

• Rate limit protects the source from excessive load. Making a scenario faster does not bypass it: parallel requests must also respect the limit.
• Rendering mode determines whether the HTTP response is enough or whether a browser is needed to run JavaScript. Not every site needs a browser, and it consumes more compute time.
• robots.txt is a separate optional setting. When enabled, a prohibited address is stopped before loading. Check the source's permission and terms for collection.
• Worker can be the platform server or your own agent. An unavailable agent can delay a run.
• Proxies are configured separately and assigned to a source through a pool.

If you created the site from a Wildberries solution, its settings include products and delivery cities. Select multiple cities to compare prices and availability: the scenario processes each city separately. Click Save settings after changing the list. In your own scenario, use the Wildberries SDK methods to obtain the selected products and cities without manually building request URLs.

Fields

Open Data to collect on the site page. Add what you want to receive, such as a product name, price, product code, and product link. Give each field a clear name and data type. Where needed, mark a field as required or as a way to identify duplicate records. Each field editor opens on the right. After making changes, click Save fields at the bottom of the page.

A field code is generated automatically. You can add a description next to the name. The editor offers value preparation, such as trimming extra spaces or parsing a price. Under Accepted values, you can set bounds, a pattern, and an allowed list. Each list item can be added or removed separately.

Under Try an example, select a prepared example or paste a value from the site. The result appears shortly after you stop typing. The pattern sits beside the sample text: start with a ready-made rule, expand a short explanation, and inspect which parts of the text matched the groups. The field code and default value are under Integration settings. You normally do not need to change them. Added fields define the result structure; collection from the site is configured separately.

Check missing prices, decimal numbers, units, and stock arrays. Do not use a product name as the sole identifier if it can change. Quality rules help detect unusual values but cannot fix an incorrect parser.

Visual mapper

The mapper opens a page through a program on your computer. Select an element, inspect its properties, and use the result to configure extraction. Close pop-ups within the page and wait for dynamic content to load. See the visual mapper guide.

Deleting a site affects related data. Save any results you need and review active integrations first.

CAPTCHA

CAPTCHA setup and API: automatic handling, opt-out, site hints, agents and verification. User guide and prices.