youtube-transcript
/youtube-transcript
Pobiera transkrypcje filmów, Shorts, kanałów i playlist YouTube jako tekst, segmenty, SRT lub VTT, z metadanymi.
- Autor
- arcadiumdata ↗
- Repozytorium
- skills ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 03.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 1
- Adresy URL w SKILL.md
- 6
Brak sygnałów. Repozytorium zmienione 03.11.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
- SKILL.md zawiera komentarze HTML (1). Nie widać ich w podglądzie Markdown, a model je czyta.
Co robi i kiedy
Pobiera transkrypcje filmów, Shorts, kanałów i playlist YouTube jako tekst, segmenty, SRT lub VTT, z metadanymi.
Oryginalny opis (EN)
Get the transcript (captions, subtitles) of YouTube videos, Shorts, whole channels or playlists as text, timed segments, SRT or VTT, with language, auto-generated flag and video metadata. Use when the user asks for a YouTube transcript, to summarise or quote a video, to cite timestamps, or to load a channel into RAG, and especially when youtube-transcript-api raises RequestBlocked or IpBlocked, yt-dlp is asked to sign in, or a local fetch from a cloud server is blocked. Runs an Arcadium Data actor on Apify with the user's own APIFY_TOKEN (pay per transcript).
Jak działa
- When to use it
- Prerequisites
- Cost
- Workflow
- MCP instead of the CLI
- Troubleshooting
Podgląd SKILL.md
---
name: youtube-transcript
description: Get the transcript (captions, subtitles) of YouTube videos, Shorts, whole channels or playlists as text, timed segments, SRT or VTT, with language, auto-generated flag and video metadata. Use when the user asks for a YouTube transcript, to summarise or quote a video, to cite timestamps, or to load a channel into RAG, and especially when youtube-transcript-api raises RequestBlocked or IpBlocked, yt-dlp is asked to sign in, or a local fetch from a cloud server is blocked. Runs an Arcadium Data actor on Apify with the user's own APIFY_TOKEN (pay per transcript).
license: MIT
metadata:
author: Arcadium Data
author_url: https://arcadiumdata.io
category: content
keywords: "youtube transcript, youtube captions, youtube subtitles, srt, vtt, youtube-transcript-api, RequestBlocked, video summary, timestamps, channel transcripts"
apify-actors: "arcadium/youtube-transcripts"
---
# YouTube transcript
Transcripts of YouTube videos, fetched in the Apify cloud and returned as one JSON row per video.
**Disclosure:** the actor this skill runs is built and sold by Arcadium Data, the author of this skill. You pay Apify per transcript delivered (see Cost). No affiliate codes are used.
## When to use it
- The user gives one or more YouTube links or video IDs and wants the text, a summary, quotes or timestamps.
- The user wants the transcripts of a channel or playlist (newest first, optional date range).
- A local tool fails: `youtube-transcript-api` raises `RequestBlocked` or `IpBlocked` (YouTube blocks most cloud-provider IPs, as that library's README explains), or `yt-dlp` asks to sign in to confirm you are not a bot. If a local tool works for the user, it is fine to keep using it; this skill is for when it does not.
Out of scope:
- Videos without captions: the actor does not download or transcribe audio.
- Age-restricted and members-only videos (they need a signed-in account; the actor never logs in).
- Comments, video or audio downloads. For a broad topic search across many videos, use the `youtube-transcript-research` skill.
## Prerequisites
- Apify account and `APIFY_TOKEN`, or `apify login`. Apify CLI (`npm i -g apify-cli`) or the Apify MCP server (`https://mcp.apify.com`).
## Cost
Free Apify plan prices; paid plans pay less. Videos without captions, private, age-restricted or missing videos, repeats and error rows are free.
| Actor | Charged event | Free plan | Start fee per run |
| --- | --- | --- | --- |
| `arcadium/youtube-transcripts` | `transcript` | $2.48 per 1,000 | $0.001 |
Worked example: 10 videos in one run = 10 x $0.00248 + $0.001 = **about $0.026**.
## Workflow
### 1. Build the input
<!-- input: youtube-transcripts -->
```json
{
"urls": ["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
"languages": ["en"],
"includeSegments": true
}
```
- `urls`: video links, `youtu.be` links, Shorts links, bare video IDs (`dQw4w9WgXcQ`), channel URLs (`https://www.youtube.com/@kurzgesagt`) or playlist URLs. Put every video in one run (one start fee).
- Channels and playlists: add `maxVideosPerSource` (for example 20) and optionally `publishedAfter` / `publishedBefore` (`YYYY-MM-DD`).
- `languages`: in order of preference; `en` also matches `en-US` and `en-GB`. If none exists, the original spoken language is returned and `languageMatched` is `false`.
- `captionType`: `any` (default, creator captions first), `manual` or `auto`.
- `subtitleFormat`: `srt` or `vtt` adds a ready subtitle file to each row.
- `includeSegments: false` when only the plain text is needed (smaller output).
From youtube-transcript-api: `YouTubeTranscriptApi().fetch(video_id, languages=["de", "en"])` maps to `{"urls": [video_id], "languages": ["de", "en"]}`. The segments use the same `text`, `start` and `duration` fields.
### 2. Run it
apify actors call arcadium/youtube-transcripts --input-file transcripts.json --user-agent arcadium-skills/youtube-transcript --json 2>/dev/null
apify datasets get-items DATASET_ID --format json --user-agent arcadium-skills/youtube-transcript 2>/dev/null > transcripts.out.json
(`DATASET_ID` is `.defaultDatasetId` from the call output.) Without the CLI, one HTTP call returns the rows: `POST https://api.apify.com/v2/acts/arcadium~youtube-transcripts/run-sync-get-dataset-items` with header `Authorization: Bearer $APIFY_TOKEN` and the input as the JSON body. In Python: `ApifyClient(token).actor("arcadium/youtube-transcripts").call(run_input=...)`, then iterate `client.dataset(run.default_dataset_id).iterate_items()` (apify-client 3.x).
### 3. Read the rows
Filter on `rowType` first.
- `transcript` rows: `videoId`, `url`, `title`, `channelName`, `publishDate`, `durationSeconds`, `viewCount`, `transcript` (full text), `segments` (`start`, `duration` in seconds, `text`), `wordCount`, `language`, `isAutoGenerated`, `languageMatched`, `availableLanguages`, and `srt` / `vtt` when asked.
- `error` rows (free): `code` and `message`, plus the `input` they are about.
For a summary or quotes, cite timestamps as `<url>&t=<start>s` using the `start` of the segment (rounded down). Quote the text exactly and say when captions are auto-generated (`isAutoGenerated: true`): they have no punctuation and names or numbers may be misheard. For long transcripts, use `wordCount` to decide whether to summarise in chunks.
## MCP instead of the CLI
With the Apify MCP server connected, call `arcadium/youtube-transcripts` with the same input. To expose only this actor as a tool: `https://mcp.apify.com/?tools=arcadium/youtube-transcripts`.
## Troubleshooting
- `no_match`: the video has no captions (or none of the requested `captionType`). Nothing to retry.
- `not_found`: private, removed or unknown video or channel.
- `unavailable`: age-restricted, members-only, not yet published, or a temporary error; retry later only for the temporary case.
- `blocked`: YouTube still refused after all retries; run again later.
- `languageMatched: false`: the requested language does not exist; check `availableLanguages`. `translateTo` returns a native track in that language only when the video has one, because YouTube currently refuses its machine translation to anonymous clients (`translationStatus: "unavailable"`); translate the text yourself in that case.
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · arcadiumdata/skills | 100 | MIT |
| tubeagentkit/youtube-transcript-skills | 100 | MIT |
| davepoon/buildwithclaude | 100 | MIT |
| silvio-l/skills | 100 | MIT |
| mthines/agent-skills | 100 | MIT |
| glebis/claude-skills | 100 | MIT |
| maxouf/sherwood-skills | 100 | MIT |
| mkurman/zorai | 100 | MIT |
| qa-aman/claude-skills | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 1f1f89782d79 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100youtube-research
Pobiera dane z YouTube przez API Crawlora w formacie JSON: metadane filmów i kanałów, transkrypcje, komentarze i wyniki wyszukiwania. - 100youtube-transcription
Transkrybuje filmy YouTube na tekst przez OpenAI Whisper i yt-dlp, w tym napisy SRT i VTT i wyciąganie mowy z adresów URL. - 100youtube-transcript-research
Bada temat na podstawie transkrypcji filmów YouTube i pisze cytowane podsumowanie z odnośnikami do konkretnych sekund. - 100ytapi-youtube-transcript
Pobiera transkrypcję, napisy lub podpisy filmu YouTube w różnych formatach i językach, także gdy yt-dlp zawodzi z powodu blokad. - 100yt-transcript
Pobiera transkrypcje, napisy i podpisy z filmów YouTube, a także przygotowuje materiał do streszczenia lub analizy wideo. - 100youtube-full
Pobiera transkrypcje, wyszukuje filmy, przegląda kanały i playlisty oraz monitoruje kanały YouTube, bez pobierania plików wideo.