YouTube Channel Scraper

The Actor reads one or many public YouTube channels and collects the sections you select — channel details, videos, Shorts, live streams, playlists, community posts, home shelves, store products, or a search performed inside the channel. Channels are accepted as URLs, @handles, legacy usernames, or UC… IDs, and are resolved to a channel ID automatically. No YouTube account, API key, or cookie is required.

Accepted input

Field	Type	Default	Description
channel_sources	array	—	Required. Channel URLs, @handles, usernames, or UC… IDs.
channel_sections	multi-select	["about","videos"]	Required. Sections to collect from every channel.
channel_search_term	string	—	Required when search_results is selected.
maxItems	integer	100	Maximum rows saved across all channels and sections. 0 removes the limit.
region_code	string	US	Two-letter country code used to localize results.
language_code	string	en	Language code for result text.

{
  "channel_sources": [
    { "url": "https://www.youtube.com/@TED" },
    { "url": "@veritasium" },
    { "url": "UC_x5XG1OV2P6uZZ5FSM9Ttw" }
  ],
  "channel_sections": ["about", "videos", "shorts"],
  "maxItems": 200,
  "region_code": "US"
}

The sections

Value	Produces	Paginates
about	One profile row per channel	—
videos	Long-form uploads, newest first	Yes
shorts	Shorts, newest first	Yes
live_streams	Past and scheduled live streams	No
playlists	Playlists the channel publishes	Yes
community_posts	Community tab posts, with attachments	Yes
home_sections	The shelves shown on the channel home tab	No
store_products	Products listed on the channel store	No
search_results	Results of channel_search_term inside the channel	Yes

Sections are collected per channel in the order listed above, and maxItems is a budget shared across all of them. Asking for about, videos, and shorts from three channels with maxItems: 100 will not necessarily reach the third channel's Shorts — raise the limit or run the sections separately when even coverage matters.

How channel targets are resolved

A UC… ID is used directly. Anything else — a URL, an @handle, or a legacy username — is sent to YouTube's own resolver first, and the resulting channel ID is written to the run log so the interpretation is visible.

Resolution costs one extra request per channel, and only for targets that are not already IDs. Supplying UC… IDs skips it entirely.

Response fields

One row per item. Which fields appear depends on the section — a profile row and a community post have little in common.

Common to every row:

Field	Contents
recordType	Section that produced the row, such as channel_about or channel_videos
source	The channel target as supplied in the input
url	Canonical URL, derived from the row's identifier
position	One-based position within the run
meta	The parent channel's profile, on paginated sections

Channel details (about):

Field	Contents
channelId, title, description	Channel identity
channelHandle	The @handle
subscriberCount, subscriberCountText	Subscriber count, numeric and displayed
videosCount, viewCount	Lifetime totals
avatar, banner	Image sets
country, joinedDate	Declared country and creation date
keywords	Channel tags
links	Public links listed on the channel

Videos, Shorts, live streams, playlists, and search results:

Field	Contents
type	video, shorts, or playlist
videoId or playlistId	Item identifier
title, description	Item text
viewCount, viewCountText	View counts
lengthText	Duration
publishedTimeText, publishedAt	Relative and ISO 8601 publication time
videoCount	Number of videos, on playlist rows
thumbnail	Thumbnail images

Community posts:

Field	Contents
postId	Post ID
contentText	Post body
authorText, authorChannelId, authorChannelHandle	Author
replyCount	Reported replies
attachment	Image, poll, or video attached to the post

{
  "recordType": "channel_about",
  "channelId": "UCAuUUnT6oDeKwE6v1NGQxug",
  "title": "TED",
  "channelHandle": "@TED",
  "subscriberCount": 27700000,
  "subscriberCountText": "27.7M",
  "videosCount": "5745",
  "viewCount": "3199315747",
  "country": "United States",
  "joinedDate": "2006-12-06",
  "url": "https://www.youtube.com/channel/UCAuUUnT6oDeKwE6v1NGQxug"
}

Behaviour on partial results

A channel that cannot be resolved is logged and skipped, and the run continues with the remaining channels. A section YouTube declines to serve for one channel is logged and skipped, and the other sections for that channel still run. Rows already collected are kept in both cases. The run fails only when no supplied target resolved to a public channel at all.

This matters on large input lists: one deleted or renamed channel among two hundred costs that channel's rows, not the run.

Promoted results carrying type: "ad" are dropped before rows are saved, so they are neither stored nor billed.

Frequently asked questions

Which forms of channel address are accepted? Full channel URLs in any form, @handles with or without the @, legacy /user/ names, and UC… IDs. Mixing forms in one run is fine. Only UC… IDs skip the resolution request.

Why did a channel return fewer videos than it has published? maxItems is shared across every channel and section in the run, so it is usually the limit that was reached first. Where a channel genuinely has more, the video and Shorts tabs paginate until YouTube stops serving pages.

Why can live streams and home shelves not be paginated? YouTube serves those tabs as a single page with no cursor to follow. The Actor collects what that page contains and moves on. Videos, Shorts, playlists, community posts, and in-channel search all paginate normally.

Can videos be sorted by popularity instead of newest? Not currently. YouTube's channel tabs accept a sort parameter but reject every value except newest, so offering the other options would only produce failed sections. Sorting the collected rows on viewCount afterwards gives the same result for a fully collected tab.

Why is store_products empty? Most channels have no store. The section returns rows only for channels that have configured one, and returns nothing rather than failing for those that have not.

How is subscriberCount different from subscriberCountText? subscriberCountText is what YouTube displays, such as 27.7M. subscriberCount is the number behind it, 27700000. YouTube rounds public subscriber counts to three significant figures above 1,000, so both are approximations — that rounding is YouTube's, not the Actor's. videosCount and viewCount are exact.

Is a YouTube account or API key required? No. Only publicly visible channel content is read, with no credentials or logged-in session, and no YouTube Data API quota is consumed.

Related Actors

Actor	Purpose
YouTube Video Scraper	Full metadata and transcripts for known videos and Shorts
YouTube Search Scraper	Keyword search, hashtags, Trending, Hype, and Home feeds
YouTube Comments Scraper	Comments from videos, Shorts, and community posts
YouTube Playlist Scraper	Playlist metadata and every video in it