Youtube transcript skill

Skill 0SxD/youtube-transcript-skill

Fetch YouTube video transcripts (auto-generated or manual subtitles) without API keys, using the open-source youtube-transcript-api Python library, with yt-dlp as a fallback for blocked videos. Use when the user asks for a transcript / captions / subtitles of a YouTube video, or when summarizing / quoting a video.From its SKILL.md

Install

npx -y skills add 0SxD/youtube-transcript-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, 498 tokens by cl100k_base, as published. Nobody here has run it

Skill: youtube-transcript

When to invoke

User mentions:

"get the transcript of this YouTube video"
"summarize this YouTube link"
"what does the speaker say at <timestamp>"
pastes a youtube.com/watch?v=... or youtu.be/... URL and asks anything that requires reading what was said

How to invoke

The skill ships a uv venv at .venv/. Run:

cd <plan-root>/skills/youtube-transcript
.venv/Scripts/python.exe fetch_transcript.py VIDEO_ID

Or call from another script:

import sys
sys.path.insert(0, "<plan-root>/skills/youtube-transcript/.venv/Lib/site-packages")
from youtube_transcript_api import YouTubeTranscriptApi
api = YouTubeTranscriptApi()
fetched = api.fetch(video_id, languages=["en"])
for s in fetched:
    print(f"[{s.start:.1f}s] {s.text}")

Fallback to yt-dlp

If youtube-transcript-api fails (geofence, rate limit, etc.):

.venv/Scripts/yt-dlp \
  --write-auto-subs --skip-download \
  --sub-langs en --sub-format vtt \
  -o "%(id)s.%(ext)s" \
  "https://www.youtube.com/watch?v=VIDEO_ID"

Returns

youtube-transcript-api: list of FetchedTranscriptSnippet objects with .text, .start, .duration
yt-dlp: a .vtt file in current directory named after the video ID

Constraints

Some videos disable transcripts entirely; both tools will fail. Tell user.
Auto-generated transcripts are imperfect; flag as "auto-generated" if is_generated=True on the transcript object.
Region-blocked videos may return 403; route through yt-dlp with --proxy if needed.

Citations

youtube-transcript-api: https://github.com/jdepoix/youtube-transcript-api (MIT)
yt-dlp: https://github.com/yt-dlp/yt-dlp (Unlicense)

Verified

Smoke-tested 2026-04-30 against video ID 0eKVizvYSUQ. Returned 128 segments. Skill works.

What ships with it: 5 files

3.7 KB alongside SKILL.md, 1 of them executable

Gives 0 of the 12 instructions most video audio skills give in 498 tokens

Counted across 622 of the 795 authors here whose files we hold, read 2026-08-07

Read individual rule files for detailed explanationsin 21 of 622, across 10 files
Render final videoin 13 of 622, across 6 files
Use WAV PCM 16kHz mono audio formatin 12 of 622, across 3 files
Use this skill when dealing with Remotion codein 11 of 622, across 4 files
Save generated audio to a WAV filein 11 of 622, across 4 files
Handle conversion errors gracefullyin 10 of 622, across 6 files
Add captions to videos alwaysin 10 of 622, across 4 files
Generate music from text descriptions using MusicGenin 9 of 622, across 2 files
Do not skip pipeline layersin 9 of 622, across 3 files
Do not make one tool do everythingin 9 of 622, across 3 files
Use Azure Document Intelligence for complex PDFsin 9 of 622, across 4 files
Never ask the user to paste their full API keyin 9 of 622, across 3 files

Said here and by no other author read

run python fetch script with video id
fallback to yt-dlp on failure
use proxy flag on region block
inform user when transcripts disabled

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.