메뉴
Use when the user asks to download, archive, search, inspect, or organize media with yt-dlp, MeTube, YouTube, playlists, channels, supported video sites, audio extraction, metadata, thumbnails, subtitles, Plex yt-dlp libraries, or NAS media downloads.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | yt-dlp |
| description | Use when the user asks to download, archive, search, inspect, or organize media with yt-dlp, MeTube, YouTube, playlists, channels, supported video sites, audio extraction, metadata, thumbnails, subtitles, Plex yt-dlp libraries, or NAS media downloads. |
Use this skill for the whole yt-dlp domain: ad hoc video downloads, playlists, channel URLs, audio-only downloads, search/preview workflows, metadata preservation, MeTube/NAS routing, and local CLI fallback.
--video. Use both mode when the user asks to grab both audio and video or explicitly passes --both.yt-dlp only when the user asks for local output, debugging, inspection, search, or a destination outside the NAS.Default for explicit URLs unless the user says local or video:
MeTube: https://metube.tootie.tv
Audio output: /mnt/user/data/media/yt-dlp-audio
Video output: /mnt/user/data/media/yt-dlp
Video Plex library: yt-dlp
Video Plex path: /data/yt-dlp
Use the slash command when possible:
/yt-dlp <url> [url ...]
/yt-dlp --video <url> [url ...]
/yt-dlp --both <url> [url ...]
MeTube is deployed on tootie and writes into the Plex-visible media share. SWAG exposes the UI at https://metube.tootie.tv.
NAS --both runs yt-dlp directly inside the metube container over SSH instead of submitting two MeTube queue jobs. This keeps audio and video archive state separate so one format does not cause the other to be skipped by MeTube's global archive.
Use local mode for filesystem-local jobs:
/yt-dlp --local <url> [url ...]
/yt-dlp --local --video <url> [url ...]
/yt-dlp --local --both <url> [url ...]
Local defaults:
audio dir: $PWD/downloads/yt-dlp-audio
video dir: $PWD/downloads/yt-dlp
archive: $PWD/downloads/.archive.txt
both-mode audio archive: $PWD/downloads/.archive-audio.txt
both-mode video archive: $PWD/downloads/.archive-video.txt
audio format: m4a
video format: bestvideo*+bestaudio/best
Environment overrides:
METUBE_URL
METUBE_BOTH_ROUTE
METUBE_CONTAINER
METUBE_QUALITY
METUBE_FORMAT
METUBE_FORMAT_SELECTOR
METUBE_AUDIO_FORMAT
METUBE_AUDIO_ARCHIVE
METUBE_VIDEO_ARCHIVE
YT_DLP_NAS_HOST
YT_DLP_DOWNLOAD_DIR
YT_DLP_ARCHIVE
YT_DLP_VIDEO_ARCHIVE
YT_DLP_FORMAT
YT_DLP_OUTPUT_TEMPLATE
YT_DLP_AUDIO_DOWNLOAD_DIR
YT_DLP_AUDIO_ARCHIVE
YT_DLP_AUDIO_FORMAT
YT_DLP_AUDIO_OUTPUT_TEMPLATE
For local video downloads, use best available quality and preserve practical metadata:
yt-dlp \
--format "bestvideo*+bestaudio/best" \
--download-archive "$archive_file" \
--output "$output_template" \
--embed-metadata \
--embed-thumbnail \
--convert-thumbnails jpg \
--write-info-json \
--write-thumbnail \
--write-description \
--write-subs \
--write-auto-subs \
--sub-langs "all,-live_chat" \
--embed-subs \
--restrict-filenames \
--yes-playlist \
--no-overwrites \
--continue \
--merge-output-format mkv
Notes:
bestvideo*+bestaudio/best chooses the best separate video/audio streams when available, falling back to the best combined stream.mkv is the safest merge container for mixed codecs and embedded subtitles. Use MP4 only when the user explicitly needs MP4 compatibility..info.json, thumbnails, descriptions) are intentional; they preserve durable metadata for later indexing or retagging.For audio-only local downloads, use --extract-audio --audio-format m4a by default. Use opus for efficient/loss-minimizing YouTube audio; use mp3 only for compatibility.
For search requests, do not download immediately. Return candidates first:
yt-dlp "ytsearch10:<query>" --flat-playlist --print "%(title)s | %(channel)s | %(duration_string)s | %(webpage_url)s"
For explicit URLs that need inspection:
yt-dlp --dump-single-json --flat-playlist "<url>"
yt-dlp --list-extractors is the local source of truth for this installation. The official supported-sites list is generated from yt-dlp extractors, but site support changes as websites change. If a site is not listed, it may still work through the generic extractor or embedded media URLs.
Common supported families include YouTube, Vimeo, SoundCloud, Bandcamp, Twitch, TikTok, Instagram, Facebook, Twitter/X, Reddit-hosted media, archive.org, Apple Podcasts, Audius, ARD/Arte/BBC-style broadcasters, and many regional TV/news sites. DRM-protected services generally do not work.
Requested format is not available: inspect with yt-dlp -F "<url>", then set YT_DLP_FORMAT.yt-dlp --version; cookies or PO-token handling may be needed for restricted content.ffmpeg: quality merging, metadata embedding, thumbnails, subtitles, and audio extraction may fail.--download-archive file before assuming a URL failed.https://metube.tootie.tv/history or the MeTube UI, then inspect the metube container logs on tootie.When downloading, report:
Use when the user wants to live-test an Android APK end-to-end on an emulator/device and get a works/doesn't-work + UI/UX report — driving the real app, not writing test code. Triggers: "test my Android app", "QA this APK", "run the app on the emulator and tell me what breaks", "click through every screen", "review the app's UX", "does my APK work", "test the built APK". Installs/launches the APK, enumerates screens from the accessibility tree, exercises each feature via adb (tap/type/swipe/launch), watches logcat for crashes/ANRs, captures screenshots + UI dumps, and emits a structured report. Primary path is direct local adb; optional richer path via claude-in-mobile. Sibling of web-app-testing and desktop-app-testing (shared report format). Does NOT fire for: building/coding an Android app (use claude-android-ninja / jetpack-compose-expert), iOS, or unit tests.
Use when the user wants to live-test a built Windows desktop application (.exe) end-to-end inside the agent-os Windows 11 VM and get a works/doesn't-work + UI/UX report — launching the real binary and driving it, not writing test code. Triggers: "test my Windows app", "QA this .exe", "run my desktop build on agent-os and tell me what breaks", "click through the app", "review the desktop app's UX", "does my exe work", "test the built binary in the Windows VM". Gets the .exe into the VM, launches it, enumerates controls from the UI Automation tree, drives every feature (click/type by element), watches for crashes/hangs/error dialogs, captures screenshots + control-tree dumps, and emits a structured report. Drives the agent-os VM via the Windows-MCP gateway. Sibling of web-app-testing and android-app-testing (shared report format). Does NOT fire for: building/coding a desktop app, the user's personal Windows on steamy (use nircmd), or general agent-os VM driving (use agent-os).
Use when the user wants to live-test a WEB app/site end-to-end in a real browser and get a works/doesn't-work + UI/UX report — not just write Playwright code. Triggers: "test my web app", "QA this site", "run E2E on the web app", "click through every feature and tell me what breaks", "review the UX of this web app", "does my web app work", "test the deployed site". Drives a real Chrome via Playwright over CDP (default http://127.0.0.1:9222), enumerates features from the DOM/ARIA tree, exercises each, captures screenshots + console/page/network errors, and emits a structured report. Sibling of android-app-testing and desktop-app-testing (shared report format). Does NOT fire for: writing a one-off Playwright script (use webwright), pure web design/build with no testing, or backend/API-only testing.
Read or interact with the broadcastr activity feed across concurrent Claude sessions. Use when the user says "what are other agents doing", "show recent activity", "mute plan-exec notifications", "what's on the bus", "broadcastr status", "tail broadcastr", "emit a manual broadcastr event", or asks why a notification appeared. Also use to disable broadcastr in the current session (BROADCASTR_DISABLED=1) or check the global feed. Does NOT fire on generic "show me activity" or "what's happening" unless broadcastr is explicitly named.
Use when the user asks to "work it", execute a superpowers executing-plans document in a worktree, create a PR as soon as the implemented plan is green, or run a complete review-and-fix loop over all touched files.
Save session documentation to a markdown file with full context — date, branch, HEAD, session ID, and git state pre-injected — then stage, commit, and push only the generated session artifact. Use when the user says "save session", "save to md", "document this session", "write up what we did", "save session notes", or asks to capture the current conversation as a session log. Pass `--html` or a `.html` path to render a rich Aurora-styled HTML artifact instead.