メニュー
Comprehensive YouTube operations using yt-dlp - download videos/audio, extract transcripts and subtitles, get metadata, work with playlists, download thumbnails, and inspect available formats. Use this for any YouTube content processing task.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | youtube |
| description | Comprehensive YouTube operations using yt-dlp - download videos/audio, extract transcripts and subtitles, get metadata, work with playlists, download thumbnails, and inspect available formats. Use this for any YouTube content processing task. |
| license | MIT |
This skill provides comprehensive YouTube operations using yt-dlp, always using the latest version via uvx.
uv - The Python package runner (provides uvx command)No pre-installation of yt-dlp is needed. The uvx command automatically fetches and runs the latest version.
Download best quality (default):
uvx yt-dlp -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Download to custom location:
uvx yt-dlp -o "/path/to/folder/%(title)s.%(ext)s" "VIDEO_URL"
Download specific quality (e.g., 720p):
uvx yt-dlp -f "bestvideo[height<=720]+bestaudio/best[height<=720]" -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
List available formats before downloading:
uvx yt-dlp -F "VIDEO_URL"
Download audio only (mp3) - downloads to Synology music folder:
uvx yt-dlp -x --audio-format mp3 -o "~/Library/CloudStorage/SynologyDrive-sync/music/%(title)s.%(ext)s" "VIDEO_URL"
Download audio in other formats:
# AAC format
uvx yt-dlp -x --audio-format aac -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
# Opus format
uvx yt-dlp -x --audio-format opus -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
# Best quality audio (no conversion)
uvx yt-dlp -f bestaudio -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Download subtitles only (no video):
# Auto-generated or manual subtitles in English
uvx yt-dlp --skip-download --write-auto-subs --sub-lang en --sub-format vtt -o "~/Downloads/%(title)s" "VIDEO_URL"
# Manual subtitles only (not auto-generated)
uvx yt-dlp --skip-download --write-subs --sub-lang en --sub-format vtt -o "~/Downloads/%(title)s" "VIDEO_URL"
Download subtitles in multiple languages:
uvx yt-dlp --skip-download --write-subs --sub-lang en,es,fr --sub-format vtt -o "~/Downloads/%(title)s" "VIDEO_URL"
List available subtitles:
uvx yt-dlp --list-subs "VIDEO_URL"
Convert subtitles to plain text (remove timestamps):
# Download as SRT first, then process
uvx yt-dlp --skip-download --write-auto-subs --sub-lang en --sub-format srt -o "~/Downloads/%(title)s" "VIDEO_URL"
# Then use sed or awk to extract just the text
sed '/^[0-9]*$/d; /^[0-9][0-9]:/d; /^$/d' ~/Downloads/video-title.en.srt > ~/Downloads/transcript.txt
Download video with embedded subtitles:
uvx yt-dlp --write-subs --embed-subs --sub-lang en -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Get video information without downloading:
# Basic info as JSON
uvx yt-dlp --dump-json "VIDEO_URL"
# Just the title
uvx yt-dlp --get-title "VIDEO_URL"
# Just the description
uvx yt-dlp --get-description "VIDEO_URL"
# Duration
uvx yt-dlp --get-duration "VIDEO_URL"
# Upload date
uvx yt-dlp --get-filename -o "%(upload_date)s" "VIDEO_URL"
# Channel/uploader
uvx yt-dlp --get-filename -o "%(uploader)s" "VIDEO_URL"
Extract all metadata to JSON file:
uvx yt-dlp --dump-json --skip-download "VIDEO_URL" > video-metadata.json
Get video chapters/timestamps:
# Chapters are included in --dump-json output
uvx yt-dlp --dump-json "VIDEO_URL" | jq '.chapters'
Download entire playlist:
uvx yt-dlp -o "~/Downloads/%(playlist)s/%(playlist_index)s-%(title)s.%(ext)s" "PLAYLIST_URL"
Download playlist as audio only:
uvx yt-dlp -x --audio-format mp3 -o "~/Library/CloudStorage/SynologyDrive-sync/music/%(playlist)s/%(title)s.%(ext)s" "PLAYLIST_URL"
Download specific videos from playlist:
# Videos 1-5
uvx yt-dlp --playlist-items 1-5 -o "~/Downloads/%(title)s.%(ext)s" "PLAYLIST_URL"
# Specific videos (1, 3, 5)
uvx yt-dlp --playlist-items 1,3,5 -o "~/Downloads/%(title)s.%(ext)s" "PLAYLIST_URL"
Get playlist information without downloading:
# Full playlist metadata
uvx yt-dlp --dump-json --flat-playlist "PLAYLIST_URL"
# Just list video titles
uvx yt-dlp --get-filename -o "%(title)s" --flat-playlist "PLAYLIST_URL"
Download only new videos from playlist (useful for subscriptions):
# Creates archive file to track downloaded videos
uvx yt-dlp --download-archive archive.txt -o "~/Downloads/%(title)s.%(ext)s" "PLAYLIST_URL"
Download thumbnail only:
uvx yt-dlp --skip-download --write-thumbnail --convert-thumbnails png -o "~/Downloads/%(title)s" "VIDEO_URL"
Download video with embedded thumbnail:
uvx yt-dlp --embed-thumbnail -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Get all available thumbnails:
uvx yt-dlp --list-thumbnails "VIDEO_URL"
Download with speed limit:
uvx yt-dlp --limit-rate 1M -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Download age-restricted content (requires cookies):
# Export cookies from browser first, then:
uvx yt-dlp --cookies cookies.txt -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Download with custom filename:
uvx yt-dlp -o "~/Downloads/my-custom-name.%(ext)s" "VIDEO_URL"
Resume interrupted download:
# yt-dlp resumes automatically if you run the same command
uvx yt-dlp -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Download with date range filter:
# Only videos uploaded after date
uvx yt-dlp --dateafter 20240101 "CHANNEL_URL"
# Only videos uploaded before date
uvx yt-dlp --datebefore 20241231 "CHANNEL_URL"
The -o flag uses templates for output filenames. Common variables:
%(title)s - Video title%(id)s - Video ID%(ext)s - File extension%(uploader)s - Channel name%(upload_date)s - Upload date (YYYYMMDD)%(playlist)s - Playlist name%(playlist_index)s - Video position in playlist%(duration)s - Video duration in seconds%(resolution)s - Video resolutionExample with multiple variables:
uvx yt-dlp -o "~/Downloads/%(uploader)s/%(upload_date)s-%(title)s.%(ext)s" "VIDEO_URL"
Create audio podcast archive from YouTube channel:
uvx yt-dlp --download-archive podcast-archive.txt -x --audio-format mp3 -o "~/Music/Podcasts/%(uploader)s/%(title)s.%(ext)s" "CHANNEL_URL"
Extract transcripts for research:
# Download all video transcripts from a playlist
uvx yt-dlp --skip-download --write-auto-subs --sub-lang en --sub-format srt -o "~/Documents/transcripts/%(title)s" "PLAYLIST_URL"
Download video with all extras:
# Video + subtitles + thumbnail + metadata
uvx yt-dlp --write-subs --embed-subs --write-thumbnail --embed-thumbnail --write-info-json -o "~/Downloads/%(title)s.%(ext)s" "VIDEO_URL"
Monitor and archive a channel:
# Run periodically to get only new uploads
uvx yt-dlp --download-archive channel-archive.txt -o "~/Videos/%(uploader)s/%(upload_date)s-%(title)s.%(ext)s" "CHANNEL_URL/videos"
~/Downloads but can be changed~/Library/CloudStorage/SynologyDrive-sync/music/ (Synology NAS)uvx ensures you always have the latest yt-dlp version without manual updatesvtt (WebVTT), srt (SubRip), json3 (timestamped JSON)--dump-json to explore all available metadata fieldsGuide for authoring Agent Skills with strong YAML `description` triggers, progressive disclosure, and bundled resources. Use when creating or updating a skill, running init_skill.py or package_skill.py, or improving a bland skill description so agents load the skill on the right user tasks.
Build an awesome marimo notebook that brings a research paper or dataset to life — for competitions, tutorials, demos, or blog companions. Use when asked to "make an awesome marimo notebook," "build a notebook for this paper," "create a competition submission," or pair-program an interactive research explainer on molab. Covers: what makes a notebook engaging (intuition game, real experiment, novel extension, design system), paper selection, narrative arc and proportions, custom anywidget patterns, wigglystuff CellTour, effective GPU usage, memory-efficient training (fused STE, gradient checkpointing), parallel review subagents, and a pre-publication polish checklist. CRITICAL: mo is auto-injected (import in ONE cell); every top-level name must be unique across ALL cells; custom anywidget needs model.save_changes(); use mo.output.replace() for live charts; guard all torch.cuda calls for CPU-only machines.
Create VS Code-style typewriter animation videos with mechanical keyboard sounds. Supports 4 typing speeds, ghost text autocomplete, strikethrough corrections, emoji picker, file tab switching, syntax highlighting (6 languages), Chinese IME (pinyin + direct mode), inline images, image stacks, animated checkboxes, Excalidraw hand-drawn diagrams, 8 visual themes, 4 sound packs, and A-roll narration sync. Built on Remotion (React-based video rendering). Use when the user wants to create code typing videos, explainer B-roll, or typewriter-style animations.
Create animated videos using Remotion from topics, product URLs, Google reviews, talking-head videos, CSV data, or channel/brand URLs. Supports 6 video types: educational explainers, product launch demos, testimonial/social proof, avatar video overlays, data visualization dashboards, and lower third overlays. Each follows a 2-step workflow: research/scrape/analyze then design and animate with spring animations, SVG diagrams, and count-up effects. Requires the Remotion best practices skill (install with `npx skills add remotion-dev/skills`). Use when the user asks to create a Remotion video, explainer video, educational video, product demo video, testimonial video, video with animated overlays, data visualization video, animated dashboard, lower third, transparent overlay, short-form vertical video for mobile, or multi-product video from a website.
Improve coherence in drafts through a four-pass sub-agent workflow: resolve argument conflicts, smooth section transitions, smooth paragraph transitions, and run a final coherence audit. Use when a user says things like 'make this coherent', 'tighten the flow', 'this feels choppy', 'can you smooth transitions', or asks to revise prose in Markdown, notes, essays, blog posts, emails, or docs so ideas connect cleanly without adding new points.
Break long-lived pull request branches into a mergeable stack of small PRs with clear dependency order and story flow. Use when a branch has grown too large, when the user asks to split a PR into stacked PRs, or when each PR must be reviewable in about 5 minutes while preserving logical narrative across the stack.