Frequently asked questions

What API can download all images from a website?

Website Image Downloader on Apify: send the page URL to POST https://api.apify.com/v2/acts/sste~website-image-downloader/run-sync-get-dataset-items and you get every image (URL, width, height, format, size, alt text); the files are in the run's images.zip. See the API docs.

How do I extract image URLs from a webpage?

Read <img src>, the largest srcset candidate, <picture> sources, lazy-load attributes (data-src, data-srcset), CSS backgrounds and og:image, and resolve them against the page URL. The guide has Python and JavaScript code; the hosted tool does it with "mode": "list".

How do I download product images in bulk?

Start from category or collection pages, keep only images above ~400 px and from the product CDN path (e.g. cdn.shopify.com), and follow product links one level deep. See bulk product images and Shopify.

How can an AI agent download website images?

Connect the MCP server https://mcp.apify.com?tools=sste/website-image-downloader to Claude, Cursor, VS Code or any MCP client. The agent gets a tool that returns image URLs, dimensions and a ZIP link. See MCP setup.

How do I filter icons and tracking pixels when scraping images?

Filter by real pixel size read from the file header (not by URL or HTML attributes): drop images under ~100×100 px, 1×1 GIFs and sprites, and exclude URLs containing logo, icon, avatar. See the guide.

How do I extract srcset images?

Split srcset into candidates (careful: URLs can contain commas), compare w/x descriptors and keep the largest; also read data-srcset and skip data: placeholders. See srcset and lazy-loaded images.

Does it work on Shopify and WooCommerce?

Yes. Product and collection pages are server-rendered; follow product links with crawlDepth: 1 and filter by the store CDN path.

Does it render JavaScript?

No, it uses fast HTTP requests. Images injected only by JavaScript after load are not seen; on static and server-rendered pages it found 96% of the images a browser shows in tests.

What does it cost?

$0.0002 per image delivered ($0.20 per 1,000), $0.0002 per page, $0.00005 per URL in list mode, $0.001 per run, plus $0.0001 per MB above 2 MB per file. Skipped and failed images are free.

Which formats?

JPEG, PNG, GIF, WebP, AVIF, SVG, ICO, BMP, TIFF and HEIC, detected from file content.

Will it bypass logins or anti-bot protection?

No. Pages that block automated access or need a login are reported as failed.

Can I use the images however I want?

Only if you have the rights. Respect copyright, licences and the terms of the websites you process.