Agent skill
omnicaptions-download
Use when downloading videos, audio, or captions from YouTube and other video platforms. Supports quality selection.
Install this agent skill to your Project
npx add-skill https://github.com/lattifai/omni-captions-skills/tree/main/skills/omnicaptions-download
SKILL.md
Download from Video Platforms
Download videos, audio, and captions from YouTube and 1000+ video platforms using yt-dlp.
Confirmation Required
IMPORTANT: Before executing any download, you MUST confirm with the user using AskUserQuestion:
- Show the URL to download
- Show the quality setting (audio/video)
- Show the output directory
- Ask for confirmation
Example confirmation:
Ready to download:
- URL: https://youtube.com/watch?v=xxx
- Type: Audio only / Video (1080p)
- Save to: Current directory
Confirm download?
Only proceed with the download command after user confirms.
When to Use
- Download YouTube videos/audio for offline use
- Extract captions from video platforms
- Get audio for local transcription or editing
When NOT to Use
- Just need transcription (use
/omnicaptions:transcribe- Gemini handles URLs directly) - Converting existing caption formats (use
/omnicaptions:convert)
Setup
pip install omni-captions-skills --extra-index-url https://lattifai.github.io/pypi/simple/
CLI Usage
Note: By default, files are saved to the current working directory. Do not specify -o unless the user explicitly requests a different location.
# Download audio only (default, saves to current directory)
omnicaptions download "https://www.youtube.com/watch?v=VIDEO_ID"
# Supports bare YouTube video ID (auto-validates via yt-dlp)
omnicaptions download e882eXLtwkI
# Download video (1080p recommended)
omnicaptions download "https://youtube.com/watch?v=VIDEO_ID" -q 1080p
# Only use -o when user explicitly requests a different location
omnicaptions download "https://youtube.com/watch?v=VIDEO_ID" -o ./downloads/
| Option | Description |
|---|---|
-o, --output |
Output directory (default: current) |
-q, --quality |
Quality: audio (default), best, 1080p, 720p, 480p, 360p |
-v, --verbose |
Verbose output |
Quality Presets
| Preset | Description |
|---|---|
audio |
Audio only (m4a/mp3), smallest size |
1080p |
1080p video + audio (recommended for video) |
720p |
720p video + audio |
480p |
480p video + audio |
360p |
360p video + audio |
best |
Best available quality (may be 4K+, very large) |
Output
Downloads produce:
- Audio/Video file:
.m4a,.mp4, etc. - Captions (if available):
.vttor.srt - Metadata:
.meta.json(video resolution, title, etc. for ASS font scaling)
Video: ./VIDEO_ID.mp4
Audio: ./VIDEO_ID.m4a
Caption: ./VIDEO_ID.en.vtt
Metadata: ./VIDEO_ID.meta.json # Used by convert for auto font size
Title: Video Title Here
The .meta.json file stores video resolution, which omnicaptions convert uses to auto-calculate font size for ASS karaoke output.
Supported Platforms
YouTube, Bilibili, Vimeo, Twitter/X, and 1000+ sites.
Related Skills
| Skill | Use When |
|---|---|
/omnicaptions:transcribe |
Transcribe downloaded audio/video |
/omnicaptions:translate |
Translate captions with Gemini |
/omnicaptions:translate |
Translate captions with Claude (no API) |
/omnicaptions:convert |
Convert caption format |
Workflow Examples
Important: Generate bilingual captions AFTER LaiCut alignment. Preserve language tag in filename.
# Has caption: download → LaiCut align (JSON) → convert → translate
omnicaptions download "https://youtube.com/watch?v=xxx"
# → xxx.en.vtt
omnicaptions LaiCut xxx.mp4 xxx.en.vtt
# → xxx.en_LaiCut.json
omnicaptions convert xxx.en_LaiCut.json -o xxx.en_LaiCut.srt
# → xxx.en_LaiCut_Claude_zh.srt (after translate)
# No caption: download → transcribe → LaiCut align (JSON) → convert → translate
omnicaptions download "https://youtube.com/watch?v=xxx"
omnicaptions transcribe xxx.mp4
# → xxx_GeminiUnd.md
omnicaptions LaiCut xxx.mp4 xxx_GeminiUnd.md
# → xxx_GeminiUnd_LaiCut.json
omnicaptions convert xxx_GeminiUnd_LaiCut.json -o xxx_GeminiUnd_LaiCut.srt
# → xxx_GeminiUnd_LaiCut_Claude_zh.srt (after translate)
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
lattifai/omni-captions-skills
omnicaptions-convert
Use when converting between caption formats (SRT, VTT, ASS, TTML, Gemini MD, etc.). Supports 30+ caption formats.
lattifai/omni-captions-skills
omnicaptions-translate
Use when translating captions/captions to another language. Supports bilingual output and context-aware translation. Default uses Claude native, Gemini API optional.
lattifai/omni-captions-skills
omnicaptions-transcribe
Use when transcribing audio/video to text with timestamps, speaker labels, and chapters. Supports YouTube URLs and local files. Produces structured markdown output.
lattifai/omni-captions-skills
omnicaptions-LaiCut
Use when user needs accurate/precise caption timing, or aligning captions with audio/video using forced alignment. Corrects caption timing to match actual speech. Uses LattifAI Lattice-1 model.
mattpocock/skills
obsidian-vault
Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.
mattpocock/skills
edit-article
Edit and improve articles by restructuring sections, improving clarity, and tightening prose. Use when user wants to edit, revise, or improve an article draft.
Didn't find tool you were looking for?