YouTube Scraper for n8n

Read YouTube data in your n8n workflows — video details, comments, transcripts, channels, playlists, search results and the trending feeds — without an API key or a quota.

The node runs Apify Actors published by thenetaji. The scraping happens on Apify's infrastructure; the rows land in your workflow as ordinary n8n items.

Looking to download a video or rip audio? That's a different package: n8n-nodes-youtube-downloader. This one reads data; that one produces files.

What you can build with it

• Summarise a video with an LLM by feeding it the transcript instead of the whole video
• Watch a competitor's channel and post new uploads to Slack
• Pull the comments on your latest video and run sentiment analysis over them
• Track what's trending in a country every morning
• Turn a playlist into a content spreadsheet

Install it

In n8n, go to Settings → Community nodes → Install and enter:

n8n-nodes-youtube-scraper-apify

See n8n's installation guide for self-hosted and Cloud specifics.

Requirements

• A self-hosted n8n, or n8n Cloud with community nodes enabled
• Node.js 20.15 or newer (self-hosted)
• A free Apify account for the API token

Connect your Apify account

1. Sign in to the Apify Console.
2. Open Settings → Integrations and copy your Personal API token.
3. In n8n, add a new Apify API credential and paste the token into API Key.

On n8n Cloud you can skip the token: set the node's Authentication to OAuth2, add an Apify OAuth2 API credential and click Connect my account.

The same credential works for every Apify node in this repo, so you only enter the token once.

Your first run

1. Add a Manual Trigger, then a YouTube Scraper node.
2. Leave Resource on Video and Operation on Get Details.
3. Paste a video URL into Videos.
4. Click Test step.

What each operation does

Video

| Operation | What comes back | | --- | --- | | Get Details | Title, views, likes, channel and metadata. Takes plain URLs or 11-character IDs. | | Get Extended Details | The same, plus optional related videos, same-sound Shorts, transcript and caption text | | Get Transcript | Timed transcript segments, optionally with a .vtt or .srt subtitle file attached | | Get Comments | One row per comment. Works on videos, Shorts and community posts. |

Get Details and Get Extended Details are two different Actors, not one Actor with a switch. Reach for Get Details by default — it is the leaner, cheaper call. Use Get Extended Details only when you actually want the related content, transcript or caption fields.

Channel, Playlist and Search

| Operation | What comes back | | --- | --- | | Channel → Get Channel | The sections you tick: channel details, videos, Shorts, live streams, playlists, community posts, home sections, store products, or a search within the channel | | Playlist → Get Playlist | One row per playlist item, optionally with full video details or transcripts attached | | Search → Search | Keyword results, filterable by type, duration, upload date and order | | Search → Search by Hashtag | A hashtag's video or Shorts feed | | Search → Get Trending | The trending, music or games chart for a country | | Search → Get Hype | The Hype chart of rising videos | | Search → Get Home Feed | The YouTube home feed for a country | | Search → Get Search Suggestions | Autocomplete suggestions for a partial query |

The six Search operations are all the same Actor with a different workflow pinned, so the node shows you only the fields that workflow actually uses.

Settings, in plain terms

Targets

Every target field takes one entry per line, and accepts commas as a separator too. Channels accept full URLs, @handles, usernames and UC… IDs. Videos accept full URLs, Shorts URLs and bare 11-character IDs.

Return All and Limit

Leave Return All off and set a Limit, or turn it on to keep going until the source runs out. Limits keep your Apify bill predictable — start small.

The "Add …" toggles

Each one attaches extra data at the cost of extra requests: Add Transcript, Add Caption Text, Add Related Content, Add Shorts Using the Same Sound, Add Full Video Details, Add Parent Content Details and Add Subtitle File. They are all off by default and are charged only when the extra data actually comes back. Turn on what you need and nothing else.

Channel Content

Get Channel requires at least one section. Channel details produces one profile record per channel; the other sections produce one row per item. Search Within Channel only appears once you tick that section, since the search term is meaningless without it.

Options

• Country and Language localize results (US, IN, DE / en, es, pt-BR). Leave them empty to take the Actor's default.
• Resume From continues a previous run — paste the resume token from that run's log. It only appears on the operations that support it.
• Actor Memory raises the memory for the run, which can shorten a large export.
• Poll Timeout is how long the node waits for the run to finish. The run keeps going on Apify's side even if the node stops waiting, and the error message links to it.

Recipes

Summarise a video

YouTube Scraper (Video → Get Transcript) → Basic LLM Chain. Feed the transcript text in as the prompt input; you get a summary without the model ever touching the video.

Comment sentiment digest

Schedule Trigger → YouTube Scraper (Video → Get Comments, Result Order = Newest Comments) → Basic LLM Chain → Slack.

Track a channel's uploads

Schedule Trigger → YouTube Scraper (Channel → Get Channel, Channel Content = Videos, Limit = 10) → Filter on publish date → Slack or Google Sheets.

Turn a playlist into a spreadsheet

YouTube Scraper (Playlist → Get Playlist) → Google Sheets → Append.

Common questions

Do I need a YouTube Data API key? No. There is no Google quota involved here.

Why did a video come back empty? Private, age-restricted, members-only and region-blocked videos return nothing. The node logs a warning with a link to the run so you can see which target failed.

Why is there no transcript? Not every video has one, and auto-generated tracks are not available in every language. The run still succeeds; the transcript field is just absent.

Can I use this as an AI tool? Yes — the node is exposed to n8n's AI Agent as a tool.

What it costs

Apify's free tier includes $5 of platform credit per month. Runs are billed by compute time, so the levers that matter are Limit and the Add … toggles. Start with small limits and the toggles off.

For developers

pnpm install          # install dependencies
pnpm run build        # compile to dist/
pnpm run lint         # n8n community-node lint
pnpm test             # unit tests for the pure helpers
pnpm run dev          # run a local n8n with this node linked

The Actor mapping lives in nodes/YouTubeScraper/operations.ts as a plain registry: each operation names its Actor, the fields it forwards, and any input it pins itself (the Search workflows). Field conversion is driven by a FIELDS table — sourceList becomes [{ url }] objects, idList stays plain strings, and optional values are omitted rather than sent blank so the Actor's own defaults apply. The node's displayOptions are derived from that registry, so the UI cannot drift out of step with the Actors.

Links

• Actors on Apify
• Repository
• Issues
• n8n community nodes

Version history

• 0.3.0 — added OAuth2 sign-in, so n8n Cloud users can connect their Apify account with a button instead of pasting a key.
• 0.2.0 — the Apify credential field is now API Key, matching the shape Apify publishes, so one saved credential works across every Apify node. Re-enter your token if you saved it under 0.1.x.
• 0.1.1 — documentation only: Apify links are now referral links.
• 0.1.0 — first release: 12 operations across Video, Channel, Playlist and Search.

Apify links in this README are referral links.

License

MIT