jxnl/youtube-extractor
agents/skills/youtube-extractor/SKILL.md
Extract transcripts, titles, and thumbnails from YouTube videos. Use for ingesting video content, capturing captions with timestamps, or downloading video metadata.
$ install --global
skillsauthnpx skillsauth add jxnl/dots youtube-extractorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 25, 2026, 6:30 AM164.9s3 files scanned
SKILL.md
- name:
- youtube-extractor
- description:
- Extract transcripts, titles, and thumbnails from YouTube videos. Use for ingesting video content, capturing captions with timestamps, or downloading video metadata.
YouTube Extractor
Overview
Use this skill to extract transcripts (with timestamps), titles, descriptions, and thumbnails from YouTube videos. Outputs are saved under a local project directory (default: ./.youtube-artifacts/<video-id>).
Quick Start
- Create a local venv and install deps:
uv venv uv add --dev yt-dlp typer - Extract all metadata and transcript:
uv run python scripts/youtube_tools.py extract "https://youtube.com/watch?v=VIDEO_ID" - Get just the transcript:
uv run python scripts/youtube_tools.py transcript "https://youtube.com/watch?v=VIDEO_ID"
Tasks
Extract all (default)
- Command:
uv run python scripts/youtube_tools.py extract <url> - Output:
./.youtube-artifacts/<video-id>/metadata.json: title, description, channel, duration, upload datetranscript.json: captions with timestampstranscript.txt: plain text transcriptthumbnail.jpg: highest resolution thumbnail
- Behavior: Downloads all available metadata and transcript.
Flags
--out-dir <dir>: output directory--no-thumbnail: skip thumbnail download--no-transcript: skip transcript extraction--lang <code>: preferred transcript language (default: en)--overwrite / --no-overwrite: overwrite existing outputs
Transcript only
- Command:
uv run python scripts/youtube_tools.py transcript <url> - Output:
transcript.jsonandtranscript.txt - Behavior: Extracts only captions/subtitles.
Flags
--out-dir <dir>: output directory--lang <code>: preferred language--format <json|txt|both>: output format (default: both)
Metadata only
- Command:
uv run python scripts/youtube_tools.py metadata <url> - Output:
metadata.json - Behavior: Extracts title, description, channel info, duration.
Thumbnail only
- Command:
uv run python scripts/youtube_tools.py thumbnail <url> - Output:
thumbnail.jpg - Behavior: Downloads highest resolution thumbnail available.
Flags
--out-dir <dir>: output directory--quality <best|high|medium|low>: thumbnail quality (default: best)
Storyboard frames
- Command:
uv run python scripts/youtube_tools.py storyboard <url> - Output:
frames/directory with individual timestamped JPGs +storyboard_manifest.json - Behavior: Extracts YouTube's preview thumbnails (used in video scrubber) into individual frames at ~2s intervals.
Flags
--out-dir <dir>: output directory--with-transcript / -t: also extract transcript and align each segment to its nearest frame--lang <code>: transcript language (if using --with-transcript)
Higher quality frames with ffmpeg
Storyboard frames are low-res (~320x180). For full quality frames aligned to transcript timestamps:
-
Download video with yt-dlp:
yt-dlp -f "bestvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]" -o video.mp4 <url> -
Extract frames at specific timestamps from transcript.json:
ffmpeg -ss 00:01:23 -i video.mp4 -frames:v 1 frame_83s.jpg -
Or extract frames at regular intervals:
ffmpeg -i video.mp4 -vf "fps=0.5" frames/frame_%04d.jpg # 1 frame every 2 seconds
Notes
- Requires network access to YouTube.
- Some videos may not have transcripts available; auto-generated captions used as fallback.
- Storyboard command requires
pillow:uv add --dev pillow
Tools
scripts/youtube_tools.py: Typer CLI withextract,transcript,metadata,thumbnail, andstoryboardcommands.
Related Skills
jxnl/yeet
tools
VerifiedTrustedCommunity
Use only when the user explicitly asks to stage, commit, push, and open a GitHub pull request in one flow using the GitHub CLI (`gh`).
182SKILL.mdUpdated May 25, 2026
jxnl/tweet-like-me
development
VerifiedTrustedCommunity
Use when Codex needs to write, rewrite, critique, or reply on Twitter/X in Jason Liu's personal voice. Trigger for requests like "tweet like me", "write this in my style", "make this sound like Jason", "draft a reply", or when Jason asks for Twitter copy about Codex, product building, feedback, launches, quote-tweets, or operator/value takes.
182SKILL.mdUpdated May 25, 2026
jxnl/simple-html-artifact
development
VerifiedTrustedCommunity
Build or refine single-file information-first HTML artifacts, especially index.html or text.html pages, with strong information hierarchy, restrained styling, accessible semantics, and minimal AI-generated frontend tells. Use when creating static HTML reports, research pages, explainers, briefs, dashboards, note indexes, or simple front ends whose goal is comprehension rather than marketing conversion.
182SKILL.mdUpdated May 25, 2026
jxnl/self-improve
development
VerifiedTrustedCommunity
Codex-specific, session-driven self-improvement for Codex behavior and project instructions. Use when the user asks to inspect past Codex sessions, run a "dream" pass over prior interactions, mine repeated user corrections/preferences, improve or draft skills, update repo/project `AGENTS.md` guidance, or propose durable edits to global `~/.codex/AGENTS.md`.
182SKILL.mdUpdated May 25, 2026