YouTube Hashtag Result Scraper

Explore public content linked to a hashtag on YouTube: pull the feed for a topic (with or without typing #), optionally narrow to Shorts only, and bias results with region and language hints. Each dataset row is one item (video or Short) with titles, channel details, view hints, thumbnails, client-style parameters when returned, and actor-added context fields.

Who it’s for

• Creators and analysts monitoring hashtag performance and related uploads.
• Researchers sampling what appears under a tag in a given market.
• Automation that stores hashtag snapshots on a schedule.

What you can do with it

• Enter a hashtag as plain text or with a leading #—the run normalizes it.
• Choose all regular videos plus Shorts, or Shorts only.
• Optional region and language hints to align the feed with an audience.
• Cap how many rows to collect across pages.

How it works (in plain terms)

You provide the hashtag and options. The actor loads the first page of results, saves rows until your cap, then automatically requests additional pages using the continuation token returned by the service, pausing briefly between requests. You do not paste tokens yourself.

Input

Field	Required	What it means
Hashtag (hashtag)	Yes	Topic text, with or without # (e.g. viral or #viral).
Content scope (contentScope)	No	Videos and Shorts (default) or Shorts only.
Region (region)	No	Country or region code to bias results (e.g. US, GB).
Language hint (language)	No	Locale or language preference (e.g. en).
Maximum results (maxResults)	No	Upper bound on rows to collect (default 100).

Output

• One row per item (until your cap). Duplicates across pages are skipped when the same videoId (or stable channel id) is seen again.
• Item kind & ids: type (e.g. shorts or video), videoId.
• Text: title, description (may be empty).
• Channel: channelTitle, channelId, channelThumbnail and channelAvatar (arrays of url / width / height), channelBadges (e.g. verification labels), isVerifiedChannel.
• Engagement & time: viewCountText, viewCount, publishedTimeText, publishDate, publishedAt, lengthText.
• Media: thumbnail (multiple sizes with url, width, height), richThumbnail (object or null), isOriginalAspectRatio.
• Playback / client tokens (when present): params, playerParams, sequenceParams.
• Actor context: hashtag (normalized tag without #), hashtagDisplay (e.g. #viral), hashtagInfoText (header summary such as video/channel counts), contentScope, optional region, scrapedAt.
• Extra fields may appear as the product changes.

Sample output (one dataset row, structure from a real run; URLs shortened)

{
  "type": "shorts",
  "videoId": "nle5f7MRUHM",
  "title": "Viral Video | Monkeys Scroll Through Social Media Like An Everyday Activity | #trending",
  "channelTitle": "CNN-News18",
  "channelId": "UCef1-8eOpJgud7szVPlZQAQ",
  "channelThumbnail": [
    {
      "url": "https://yt3.ggpht.com/…/s68-c-k-c0x00ffffff-no-rj",
      "width": 68,
      "height": 68
    }
  ],
  "channelAvatar": [
    {
      "url": "https://yt3.ggpht.com/…/s68-c-k-c0x00ffffff-no-rj",
      "width": 68,
      "height": 68
    }
  ],
  "channelBadges": ["Verified"],
  "isVerifiedChannel": true,
  "description": "",
  "viewCountText": "798,905 views",
  "viewCount": "798905",
  "publishedTimeText": "3 years ago",
  "publishDate": "2023-03-26",
  "publishedAt": "2023-03-26T00:00:00Z",
  "lengthText": "0:09",
  "thumbnail": [
    {
      "url": "https://i.ytimg.com/vi/nle5f7MRUHM/hq2.jpg?…",
      "width": 168,
      "height": 94
    }
  ],
  "richThumbnail": null,
  "isOriginalAspectRatio": true,
  "params": "CBUwAroBAA%3D%3D",
  "playerParams": "8AEBoAMFyAMkuAQVogYVATtgMiL6hNhlruC4k_qBcz3fOFW3kAcC",
  "sequenceParams": "CgtubGU1ZjdNUlVITSoCGBU%3D",
  "hashtag": "viral",
  "contentScope": "all",
  "hashtagDisplay": "#viral",
  "hashtagInfoText": "736M videos • 48M channels",
  "region": "US",
  "scrapedAt": "2026-03-27T01:26:49.296Z"
}

The real thumbnail array usually contains several entries with different dimensions; only one is shown above.