YouTube Channel API

The FetchTranscript YouTube Channel API has two endpoints. The first takes a channel’s @handle or UC… ID and returns its permanent ID, name, description, subscriber count, avatar and URL. The second lists the channel’s videos page by page, up to 100 at a time, with title, duration, views and thumbnail, Shorts optional. Together with the transcript endpoint they turn a channel name into the full text of everything it has published.

No credit card needed. One API key works on every endpoint.

Quick example

GET/v1/channels/{channel_id}
GET/v1/channels/{channel_id}/videos?limit=30&offset=0&include_shorts=false

Both calls take the same channel identifier in the path and your API key in the X-API-Key header.

# Channel details
curl "https://api.fetchtranscript.com/v1/channels/@veritasium" \
  -H "X-API-Key: yt_your_api_key"

# Its videos, 50 per page
curl "https://api.fetchtranscript.com/v1/channels/@veritasium/videos?limit=50&offset=0" \
  -H "X-API-Key: yt_your_api_key"

Sample responses

Channel details

{
  "channel_id": "UCHnyfMqiRRG1u-2MsSQLbXA",
  "name": "Veritasium",
  "description": "An element of truth - videos about science, education, and anything else we find interesting.",
  "subscriber_count": 21500000,
  "video_count": null,
  "thumbnail_url": "https://yt3.googleusercontent.com/7vCbvtCqtjQ3YLgsJt7Y952MQV1sBvhllSCSxHP8_sVZdcPCBrITfhkN2RdyCuwPnsByq-1GoA=s0",
  "channel_url": "https://www.youtube.com/channel/UCHnyfMqiRRG1u-2MsSQLbXA"
}
FieldTypeDescription
channel_idstringThe permanent channel ID (UC…), even when you asked by @handle. Store this one: handles can change.
namestringThe channel name.
descriptionstringThe “About” text of the channel.
subscriber_countinteger | nullSubscribers as YouTube displays them (rounded); null when the channel hides the count.
video_countinteger | nullNumber of videos when YouTube reports it, otherwise null.
thumbnail_urlstring | nullURL of the channel’s thumbnail image.
channel_urlstringThe channel page on YouTube.

Channel videos

The second page of two videos, requested with limit=2&offset=2.

{
  "channel_id": "UCHnyfMqiRRG1u-2MsSQLbXA",
  "videos": [
    {
      "video_id": "eyZEqSPeMwg",
      "title": "Relativity is so much weirder than we thought...",
      "description": null,
      "thumbnail_url": "https://i.ytimg.com/vi/eyZEqSPeMwg/hq720.jpg",
      "duration": 2068,
      "view_count": 8500000
    },
    {
      "video_id": "NiCgcUTL9_Q",
      "title": "Fiber Optics are so much cooler than I thought",
      "description": null,
      "thumbnail_url": "https://i.ytimg.com/vi/NiCgcUTL9_Q/hq720.jpg",
      "duration": 2072,
      "view_count": 16000000
    }
  ],
  "total_count": null,
  "offset": 2,
  "limit": 2,
  "has_more": true
}

Illustrative values: a real response carries the live data of the video or channel you ask for.

FieldTypeDescription
channel_idstringThe channel ID (UC…).
videos[].video_idstringThe video ID, ready for the transcript, comments and metadata endpoints.
videos[].titlestringThe video title.
videos[].descriptionstring | nullA description snippet when the channel page shows one, usually null.
videos[].thumbnail_urlstring | nullURL of the video thumbnail.
videos[].durationinteger | nullLength in seconds.
videos[].view_countinteger | nullViews as the channel page rounds them (8.5M becomes 8500000).
total_countinteger | nullTotal number of videos in the list when known, otherwise null.
offset, limitintegerThe page you asked for.
has_morebooleanTrue when there is another page: add the number of videos you received to offset and ask again.

Parameters

ParameterInTypeDefaultDescription
channel_idpathstringrequiredThe channel’s @handle (the @ is optional) or its UC… ID of 24 characters. Used by both endpoints.
limitqueryinteger30Videos per page, from 1 to 100. Video list only.
offsetqueryinteger0How many videos to skip, for pagination. Video list only.
include_shortsquerybooleanfalseAdd the channel’s Shorts to the list. Video list only.

An identifier that is neither a handle nor a channel ID is refused with 400 invalid_channel_id, a channel that does not exist gets 404 channel_not_found, and a limit or offset out of range gets a 422 validation error. None of these is billed.

Pagination, ordering and Shorts

Without Shorts, the list follows the channel’s Videos tab: newest first, limit videos starting at offset. When YouTube reports the total, total_count holds it and has_more compares it with your position; when it does not, has_more is true whenever the page came back full, and the last page is the first one that comes back short.

With include_shorts=true the API reads the channel’s videos and Shorts together, sorts them by view count, most viewed first, and sets total_count to the combined number. Because it reads the whole channel, that request takes longer on channels with thousands of uploads. Channel details are cached for 24 hours and video lists for one hour.

Transcripts of a whole channel

The most requested workflow: give it a channel, get back the transcript of every video. List the videos with the SDK helper, which pages through the channel 50 videos at a time, then fetch each transcript. Catch the error of videos without captions and move on.

from fetchtranscript import FetchTranscriptClient, VideoNotFoundError

client = FetchTranscriptClient(api_key="yt_your_api_key")

# Pages of 50 videos, newest first, until has_more is false: 4 calls for 200 videos
videos = client.channels.get_all_videos("@veritasium", max_videos=200)

transcripts = {}
for video in videos:
    try:
        transcripts[video["video_id"]] = client.transcripts.get_text(video["video_id"])
    except VideoNotFoundError as error:
        # No captions on this video: the failed call is not billed
        print("skipped", video["video_id"], error)

print(len(transcripts), "transcripts out of", len(videos), "videos")
import { FetchTranscriptClient, VideoNotFoundError } from 'fetchtranscript';

const client = new FetchTranscriptClient({ apiKey: 'yt_your_api_key' });

const videos = await client.channels.getAllVideos('@veritasium', { maxVideos: 200 });

const transcripts = new Map<string, string>();
for (const video of videos) {
  try {
    transcripts.set(video.video_id, await client.transcripts.getText(video.video_id));
  } catch (error) {
    if (!(error instanceof VideoNotFoundError)) throw error; // no captions: skip it
  }
}

For a channel with 200 videos this is 4 calls to list them and at most 200 transcript calls, 204 credits in all, about $0.20 at pay-as-you-go prices. The rate limit of 500 requests per minute leaves room to fetch transcripts in parallel; for very large channels, the bulk packs lower the price per call.

Use cases

Transcripts of a whole channel

List every video of a channel and download all its transcripts: a searchable archive of a podcast, a course or a creator’s back catalogue, or a knowledge base for a chatbot.

Creator and influencer research

Compare channels by subscribers, publishing rhythm and views per video before a sponsorship, a partnership or an outreach campaign.

Competitor monitoring

Poll the first page of a competitor’s videos every day and get notified when they publish, with the title and duration of each new upload.

Channel analytics dashboards

Track public numbers over time for the channels you care about: subscriber count, the views of recent videos, and how Shorts perform against long-form videos.

Resolving handles to IDs

Turn the @handles your users paste into permanent UC… channel IDs, and keep your database stable when a creator renames their handle.

Content catalogues

Build a browsable catalogue of a channel’s videos for an education platform, a fan site or an internal library, with thumbnails and durations.

Pricing

Channel details cost one credit, and so does each page of up to 100 videos: listing a channel of 1,000 videos takes 10 calls.

  • 25 free calls every month, no credit card
  • Then $1 per 1,000 calls, pay as you go
  • Only successful calls are billed: errors cost nothing
  • Credits never expire; bulk packs from 10,000 calls for $9

See all prices and bulk packs →

Also in the MCP server for AI assistants

The FetchTranscript MCP server gives Claude, Cursor and other MCP clients the same data as tools: get_channel_info, list_channel_videos and get_transcript. Ask “what has this channel published lately?” and the assistant calls get_channel_info and list_channel_videos, then reads the videos that matter with get_transcript.

Run it with uvx fetchtranscript-mcp and your API key in FETCHTRANSCRIPT_API_KEY; the MCP server guide has the configuration for each client. Tool calls are billed like API calls.

YouTube Channel API: frequently asked questions

Can I look up a channel by its @handle?
Yes. Both endpoints accept the @handle (with or without the @) or the UC… channel ID. The response always contains the UC… ID, which does not change when the creator renames the handle.
How do I list all the videos of a channel?
Call GET /v1/channels/{channel_id}/videos with limit (up to 100) and offset, and keep adding the number of videos you received to offset while has_more is true. The Python and TypeScript SDKs do this for you with get_all_videos and getAllVideos.
Are YouTube Shorts included?
Not by default: the list covers the channel’s regular videos, newest first. Add include_shorts=true to include Shorts; the combined list is then sorted by view count, most viewed first.
Does the video list include publish dates?
No. Each item has the video ID, title, thumbnail, duration and view count. Call the video metadata endpoint for the upload date, exact view count, likes and tags of a video.
How do I get the transcripts of a whole channel?
List the channel’s videos, then call GET /v1/transcripts/{video_id} for each one. For 200 videos that is 4 list calls of 50 plus 200 transcript calls, about $0.20; videos without captions return an error and are not billed.
How much does the YouTube Channel API cost?
One credit per successful call: one for the channel details, one for each page of up to 100 videos. Every account gets 25 free calls per month, then it is $1 per 1,000 calls.

Get a channel’s videos and transcripts

Create an account and get 25 free calls every month, enough to try the whole-channel workflow on a few videos.