baoyu-youtube-transcript

Name: baoyu-youtube-transcript
Author: jimliu

byjimliu· 68 skills

Installs

18k

Stars

24k

Forks

2.7k

How to install baoyu-youtube-transcript?

jimliu/baoyu-skills/baoyu-youtube-transcript

$npx -y skills add jimliu/baoyu-skills --skill baoyu-youtube-transcript

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Use this skill

Run `npx skills use "https://github.com/jimliu/baoyu-skills" --skill "jimliu/baoyu-skills/baoyu-youtube-transcript"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the whole pack

Use the skills in "https://github.com/jimliu/baoyu-skills" that are relevant to the current task. Run `npx skills add "https://github.com/jimliu/baoyu-skills"` and select the relevant skills, then follow their instructions.

Files · 1

View on GitHub

SKILL.md

1	# YouTube Transcript
2
3	Downloads transcripts (subtitles/captions) from YouTube videos. Works with both manually created and auto-generated transcripts. No API key or browser required — uses YouTube's InnerTube API directly and automatically falls back to `yt-dlp` when YouTube blocks the direct API path.
4
5	Fetches video metadata and cover image on first run, caches raw data for fast re-formatting.
6
7	## Script Directory
8
9	Scripts in `scripts/` subdirectory. `{baseDir}` = this SKILL.md's directory path. Resolve `${BUN_X}` runtime: if `bun` installed → `bun`; if `npx` available → `npx -y bun`; else suggest installing bun. Replace `{baseDir}` and `${BUN_X}` with actual values.
10
11	\| Script \| Purpose \|
12	\|--------\|---------\|
13	\| `scripts/main.ts` \| Transcript download CLI \|
14
15	## Usage
16
17	```bash
18	# Default: markdown with timestamps (English)
19	${BUN_X} {baseDir}/scripts/main.ts <youtube-url-or-id>
20
21	# Specify languages (priority order)
22	${BUN_X} {baseDir}/scripts/main.ts <url> --languages zh,en,ja
23
24	# Without timestamps
25	${BUN_X} {baseDir}/scripts/main.ts <url> --no-timestamps
26
27	# With chapter segmentation
28	${BUN_X} {baseDir}/scripts/main.ts <url> --chapters
29
30	# With speaker identification (requires AI post-processing)
31	${BUN_X} {baseDir}/scripts/main.ts <url> --speakers
32
33	# SRT subtitle file
34	${BUN_X} {baseDir}/scripts/main.ts <url> --format srt
35
36	# Translate transcript
37	${BUN_X} {baseDir}/scripts/main.ts <url> --translate zh-Hans
38
39	# List available transcripts
40	${BUN_X} {baseDir}/scripts/main.ts <url> --list
41
42	# Force re-fetch (ignore cache)
43	${BUN_X} {baseDir}/scripts/main.ts <url> --refresh
44	```
45
46	## Options
47
48	\| Option \| Description \| Default \|
49	\|--------\|-------------\|---------\|
50	\| `<url-or-id>` \| YouTube URL or video ID (multiple allowed) \| Required \|
51	\| `--languages <codes>` \| Language codes, comma-separated, in priority order \| `en` \|
52	\| `--format <fmt>` \| Output format: `text`, `srt` \| `text` \|
53	\| `--translate <code>` \| Translate to specified language code \| \|
54	\| `--list` \| List available transcripts instead of fetching \| \|
55	\| `--timestamps` \| Include `[HH:MM:SS → HH:MM:SS]` timestamps per paragraph \| on \|
56	\| `--no-timestamps` \| Disable timestamps \| \|
57	\| `--chapters` \| Chapter segmentation from video description \| \|
58	\| `--speakers` \| Raw transcript with metadata for speaker identification \| \|
59	\| `--exclude-generated` \| Skip auto-generated transcripts \| \|
60	\| `--exclude-manually-created` \| Skip manually created transcripts \| \|
61	\| `--refresh` \| Force re-fetch, ignore cached data \| \|
62	\| `-o, --output <path>` \| Save to specific file path \| auto-generated \|
63	\| `--output-dir <dir>` \| Base output directory \| `youtube-transcript` \|
64
65	## Optional Environment Variables
66
67	\| Variable \| Description \|
68	\|----------\|-------------\|
69	\| `YOUTUBE_TRANSCRIPT_COOKIES_FROM_BROWSER` \| Passed to `yt-dlp --cookies-from-browser` during fallback, e.g. `chrome`, `safari`, `firefox`, or `chrome:Profile 1` \|
70
71	## Input Formats
72
73	Accepts any of these as video input:
74	- Full URL: `https://www.youtube.com/watch?v=dQw4w9WgXcQ`
75	- Short URL: `https://youtu.be/dQw4w9WgXcQ`
76	- Embed URL: `https://www.youtube.com/embed/dQw4w9WgXcQ`
77	- Shorts URL: `https://www.youtube.com/shorts/dQw4w9WgXcQ`
78	- Video ID: `dQw4w9WgXcQ`
79
80	## Output Formats
81
82	\| Format \| Extension \| Description \|
83	\|--------\|-----------\|-------------\|
84	\| `text` \| `.md` \| Markdown with frontmatter (incl. `description`), title heading, summary, optional TOC/cover/timestamps/chapters/speakers \|
85	\| `srt` \| `.srt` \| SubRip subtitle format for video players \|
86
87	## Output Directory
88
89	```
90	youtube-transcript/
91	├── .index.json # Video ID → directory path mapping (for cache lookup)
92	└── {channel-slug}/{title-full-slug}/
93	├── meta.json # Video metadata (title, channel, description, duration, chapters, etc.)
94	├── transcript-raw.json # Raw transcript snippets from YouTube API (cached)
95	├── transcript-sentences.json # Sentence-segmented transcript (split by punctuation, merged across snippets)
96	├── imgs/
97	│ └── cover.jpg # Video thumbnail
98	├── transcript.md # Markdown transcript (generated from sentences)
99	└── transcript.srt # SRT subtitle (generated from raw snippets, if --format srt)
100	```
101
102	- `{channel-slug}`: Channel name in kebab-case
103	- `{title-full-slug}`: Full video title in kebab-case
104
105	The `--list` mode outputs to stdout only (no file saved).
106
107	## Caching
108
109	On first fetch, the script saves:
110	- `meta.json` — video metadata, chapters, cover image path, language info
111	- `transcript-raw.json` — raw transcript snippets from YouTube API (`{ text, start, duration }[]`)
112	- `transcript-sentences.json` — sentence-segmented transcript (`{ text, start: "HH:mm:ss", end: "HH:mm:ss" }[]`), split by sentence-ending punctuation (`.?!…。？！` etc.), timestamps proportionally allocated by character length, CJK-aware text merging
113	- `imgs/cover.jpg` — video thumbnail
114
115	Subsequent runs for the same video use cached data (no network calls). Use `--refresh` to force re-fetch. If a different language is requested, the cache is automatically refreshed.
116
117	When YouTube returns anti-bot / blocked responses on the direct InnerTube path, the script retries with alternate client identities and then falls back to `yt-dlp` if available. If fallback is needed but `yt-dlp` is unavailable, the agent should decide how to make `yt-dlp` available and continue rather than pushing the installation decision to the user.
118
119	SRT output (`--format srt`) is generated from `transcript-raw.json`. Text/markdown output uses `transcript-sentences.json` for natural sentence boundaries.
120
121	## Workflow
122
123	When user provides a YouTube URL and wants the transcript:
124
125	1. Run with `--list` first if the user hasn't specified a language, to show available options
126	2. Always single-quote the URL when running the script — zsh treats `?` as a glob wildcard, so an unquoted YouTube URL causes "no matches found": use `'https://www.youtube.com/watch?v=ID'`
127	3. Default: run with `--chapters --speakers` for the richest output (chapters + speaker identification)
128	3. The script auto-saves cached data + output file and prints the file path
129	4. For `--speakers` mode: after the script saves the raw file, follow the speaker identification workflow below to post-process with speaker labels
130
131	When user only wants a cover image or metadata, running the script with any option will also cache `meta.json` and `imgs/cover.jpg`.
132
133	When re-formatting the same video (e.g., first text then SRT), the cached data is reused — no re-fetch needed.
134
135	## Chapter & Speaker Workflow
136
137	### Chapters (`--chapters`)
138
139	The script parses chapter timestamps from the video description (e.g., `0:00 Introduction`), segments the transcript by chapter boundaries, groups snippets into readable paragraphs, and saves as `.md` with a Table of Contents. No further processing needed.
140
141	If no chapter timestamps exist in the description, the transcript is output as grouped paragraphs without chapter headings.
142
143	### Speaker Identification (`--speakers`)
144
145	Speaker identification requires AI processing. The script outputs a raw `.md` file containing:
146	- YAML frontmatter with video metadata (title, channel, date, cover, description, language)
147	- Video description (for speaker name extraction)
148	- Chapter list from description (if available)
149	- Raw transcript in SRT format (pre-computed start/end timestamps, token-efficient)
150
151	After the script saves the raw file, spawn a sub-agent (use a cheaper model like Sonnet for cost efficiency) to process speaker identification:
152
153	1. Read the saved `.md` file
154	2. Read the prompt template at `{baseDir}/prompts/speaker-transcript.md`
155	3. Process the raw transcript following the prompt:
156	- Identify speakers using video metadata (title → guest, channel → host, description → names)
157	- Detect speaker turns from conversation flow, question-answer patterns, and contextual cues
158	- Segment into chapters (use description chapters if available, else create from topic shifts)
159	- Format with `Speaker Name:` labels, paragraph grouping (2-4 sentences), and `[HH:MM:SS → HH:MM:SS]` timestamps
160	4. Overwrite the `.md` file with the processed transcript (keep the YAML frontmatter)
161
162	When `--speakers` is used, `--chapters` is implied — the processed output always includes chapter segmentation.
163
164	## Error Cases
165
166	\| Error \| Meaning \|
167	\|-------\|---------\|
168	\| Transcripts disabled \| Video has no captions at all \|
169	\| No transcript found \| Requested language not available \|
170	\| Video unavailable \| Video deleted, private, or region-locked \|
171	\| IP blocked \| Too many requests, try again later \|
172	\| Age restricted \| Video requires login for age verification \|
173	\| bot detected \| The script retries alternate clients and then `yt-dlp`; if fallback tooling is missing, the agent should resolve that itself, otherwise if it still fails try `YOUTUBE_TRANSCRIPT_COOKIES_FROM_BROWSER=safari` (or your browser) \|

6 picks

Type

Run `npx skills use "https://github.com/jimliu/baoyu-skills" --skill "jimliu/baoyu-skills/baoyu-youtube-transcript"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the skills in "https://github.com/jimliu/baoyu-skills" that are relevant to the current task. Run `npx skills add "https://github.com/jimliu/baoyu-skills"` and select the relevant skills, then follow their instructions.

baoyu-youtube-transcript

How to install baoyu-youtube-transcript?

Related

baoyu-youtube-transcript

How to install baoyu-youtube-transcript?

Related