Bulk Image Downloader

Need every product photo, gallery image or asset from a page as one download instead of dozens of right-clicks? Point this at a page or a list of direct image URLs and get back a single ZIP file with every image that passes your size and type filters.

Get a clean ZIP archive plus a per-image status report (kept, skipped, and why) as JSON, CSV or Excel, or call it as a tool from Claude, Cursor, ChatGPT or any MCP client. Pay only for the images actually included in the ZIP.

What you get

For each run: a single ZIP file in the run's key-value store containing every image that was downloaded and passed your filters, plus a dataset with one status row per URL checked (included, reason when skipped, content type, byte size, and the filename it got in the ZIP). Give it a page URL to scan for <img> tags, a list of direct image URLs, or both.

Every status row is a flat record with stable field names, so it drops straight into a spreadsheet, a database or an AI agent's context, while the ZIP itself is the deliverable most buyers actually want.

Why use this instead of downloading by hand

• Scans a page's <img> tags and srcset attributes for you, or takes a URL list directly
• Filters by content type and min/max file size before you pay for anything
• One ZIP file out, not dozens of individual downloads
• Works as an MCP tool, so AI agents can fetch a page's images on demand
• No browser, no proxies, no personal data: fast runs and a predictable cost per image

Input

Field	Type	Default	Meaning
startUrl	string		A page to scan for images (<img> tags and srcset). Optional if you pass urls directly.
urls	array		Specific image URLs to download, instead of or alongside startUrl.
contentTypes	array		Keep only images whose content-type contains one of these, e.g. ["png","webp"]. Empty allows all image types.
minBytes	integer	0	Skip images smaller than this many bytes.
maxBytes	integer	10000000	Skip images larger than this many bytes. Keeps the ZIP and run memory bounded.
maxResults	integer	50	Cap on images downloaded and zipped. You are charged per image included, so this also caps your cost.

Example input:

{
  "startUrl": "https://en.wikipedia.org/wiki/Solar_System",
  "maxResults": 12,
  "maxBytes": 5000000
}

Output

The ZIP: after the run finishes, open its Storage > Key-value store tab in the Apify Console and download the record named IMAGES (or fetch it directly, see below). The run's OUTPUT record also lists the direct API URL and how many images were included.

Per-URL status row (dataset), one per image candidate checked:

{
  "url": "https://upload.wikimedia.org/wikipedia/commons/thumb/example.jpg",
  "included": true,
  "reason": null,
  "contentType": "image/jpeg",
  "sizeBytes": 184213,
  "filename": "example.jpg"
}

Field reference: url (the image URL checked), included (true if it made it into the ZIP), reason (why it was skipped, e.g. "HTTP 404", "not an image content-type", "filtered by contentTypes", "larger than maxBytes (...)", null when included), contentType, sizeBytes, filename (its name inside the ZIP, de-duplicated automatically when two images share a name).

Fetch the ZIP directly once you have a run's key-value store ID (also printed in OUTPUT):

$curl -o images.zip "https://api.apify.com/v2/key-value-stores/<storeId>/records/IMAGES?token=YOUR_APIFY_TOKEN"

Pricing

Pay per event. You are charged $8.00 per 1,000 images actually downloaded and included in the ZIP, plus a fraction of a cent per run start. Nothing is charged for images that are skipped by your filters or fail to download. Set "Max total charge per run" in the run options to cap spending on any run. When a run reaches your cap it stops cleanly and keeps everything it already zipped.

Rough guide: 1,000 images cost $8.00 and take a while longer than a metadata-only Actor, since actual image bytes are downloaded; a typical run of a few dozen images finishes in well under a minute.

Use it from an AI agent (MCP)

This Actor is available as an MCP tool through the Apify MCP server. Add it to your client, then ask the agent for the data in plain language.

Claude Desktop, Claude Code or Cursor (mcp.json / claude_desktop_config.json):

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com/?actors=brightpath-data/bulk-image-downloader",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

ChatGPT and other clients that support remote MCP servers: add https://mcp.apify.com/?actors=brightpath-data/bulk-image-downloader as a connector with your Apify token.

Example prompt once connected: "Download every product image on this page as a zip: https://example.com/catalog"

Use it from code

curl -X POST "https://api.apify.com/v2/acts/brightpath-data~bulk-image-downloader/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"startUrl":"https://en.wikipedia.org/wiki/Solar_System","maxResults":12,"maxBytes":5000000}'

Python (grab the ZIP after the run finishes):

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("brightpath-data/bulk-image-downloader").call(run_input={"startUrl": "https://en.wikipedia.org/wiki/Solar_System", "maxResults": 12})
store = client.key_value_store(run["defaultKeyValueStoreId"])
with open("images.zip", "wb") as f:
    f.write(store.get_record("IMAGES")["value"])

Limits and fair use

• Up to 2,000 images per run; images are heavier than typical dataset rows, so very large runs take longer and use more memory. maxBytes (default 10 MB per image) keeps any single run bounded.
• Only URLs whose response content-type starts with image/ are downloaded; everything else is skipped and reported with a reason.
• One request per image, with a short pause between requests to stay polite to the source server; a site that blocks automated requests will show up as failed rows with the HTTP status it returned, rather than being bypassed.
• The scan only reads <img src> and the first URL in each srcset; images injected purely by client-side JavaScript after page load are not seen (no browser is used).

Data source and legal

This Actor downloads only images that are already publicly served by the page or URLs you give it, the same way your own browser would when loading that page. It does not bypass any login, paywall or access control. You are responsible for having the right to download and use the images you request.

Support

Found a problem or need a field added? Open an issue on the Actor's Issues tab. Fixes for broken runs are prioritized.