agentleFS
AtmoskunAtmoskun/claude-skills/skills/youtube-downloader/SKILL.md

youtube-downloader

Skill0 starsChanged 9 months agoMITView on GitHub

Download YouTube channel transcripts with batch support and state management. Scans channels for new videos, downloads transcripts as Obsidian-compatible Markdown. Use when the user invokes /youtube-downloader followed by a subcommand (list, check, download).

37 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use agent-deep-links, helium-mcp and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ComposioHQSkill

    agent-deep-links

    Build, validate, and troubleshoot deep links for Codex, Cursor, VS Code, Visual Studio, and similar tools. Use when users ask for clickable links (especially in Slack) that open threads, files, folders, or app settings.

    Also in 3 projects that use this

  • ComposioHQSkill

    helium-mcp

    Search real-time news with bias scoring, get live stock/ETF/crypto data with AI analysis, ML options pricing, balanced news synthesis, and meme search via the Helium MCP server.

    Also in 3 projects that use this

  • trendinuxSkill

    twitter-algorithm-optimizer

    Analyze and optimize tweets for maximum reach using Twitter's open-source algorithm insights. Rewrite and edit user tweets to improve engagement and visibility based on how the recommendation system ranks content.

    Also in 12 projects that use this

  • petestewartSkill

    datadog-logs

    Query Datadog logs via the Logs Search API v2. Use when the user asks to "check the logs", "search logs", "find errors in Datadog", "look up logs for", "show me production logs", "check for errors", "search Datadog", or any request involving log search, error investigation, or production debugging that requires Datadog log data.

    Also in 3 projects that use this

  • MicrosoftSkill

    deploy-pipeline

    Triggers a Power Platform Pipeline deployment run for a Power Pages solution. Selects a target stage, validates the package, optionally configures deployment settings (environment variables, connection references), then deploys and polls for completion. Use when asked to: "deploy pipeline", "run pipeline", "trigger deployment", "deploy to staging", "deploy to production", "run power platform pipeline", "deploy solution via pipeline", "promote solution", "push to staging", "push to production".

    Also in 3 projects that use this

  • +7 more from ComposioHQ

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 3 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 35 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read Start/Youtube Channels.md, parse the markdown table for channel entries.

    line 28
  2. Distinctive:

    If no state file exists for a channel, show all zeros and note "(not yet scanned)".

    line 40
  3. Very specific:

    Default: check = check -7d (last 7 days).

    line 44
  4. Very specific:

    Read Start/Youtube Channels.md, parse table.

    line 46
  5. Very specific:

    Default: download = download -7d (scan last 7 days, then download pending).

    line 74
  6. Very specific:

    Run check -{x}d workflow first (scan + update state with date filter.

    line 76
  7. Distinctive:

    When complete, report download summary per channel:

    line 82
  8. Very specific:

    Always read Start/Youtube Channels.md fresh at invocation start.

    line 125
  9. Distinctive:

    No API keys needed — both yt-dlp and youtube-transcript-api work without authentication.

    line 129

“Common” is measured: kept by at least 6 in 10 of the 35 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

YouTube Downloader Skill

Scope

  • Channel config: C:\Users\10444\Desktop\Claude Download\Start\Youtube Channels.md
  • State files: C:\Users\10444\.claude\skills\youtube-downloader\data\{handle}.json
  • Output: C:\Users\10444\Desktop\Claude Download\Info-Lake\Youtube-transcript\{Channel Name}\
  • Script: python _shared/run.py youtube_client.py
  • Runner CWD: C:\Users\10444\.claude\skills\

Command Format

  • /youtube-downloader list — Show channel status summary
  • /youtube-downloader check — Scan with date filter (default: last 7 days)
  • /youtube-downloader check -{x}d — Scan with date filter (last x days)
  • /youtube-downloader check -new — Scan for new videos without date filter (flat, default limit 21)
  • /youtube-downloader download — Scan last 7 days + download pending (default)
  • /youtube-downloader download -{x}d — Scan last x days + download pending
  • /youtube-downloader download -new — Download pending transcripts (no scan, default max 21)

Workflow: list

  1. Read Start/Youtube Channels.md, parse the markdown table for channel entries.
  2. For each channel, load its state JSON from youtube-downloader/data/{handle_without_@}.json.
  3. Display a summary table:
YouTube Channels Status
=======================
| Channel          | Total | Downloaded | Pending | Failed | Unavailable |
|------------------|-------|------------|---------|--------|-------------|
| All-In Podcast   |    42 |         38 |       2 |      1 |           1 |

If no state file exists for a channel, show all zeros and note "(not yet scanned)".

Workflow: check / check -{x}d

Default: check = check -7d (last 7 days).

  1. Read Start/Youtube Channels.md, parse table.
  2. For each channel, run:
cd "C:/Users/10444/.claude/skills" && python _shared/run.py youtube_client.py scan --channel-url "<url>" --channel-name "<Channel Name>" --days {x} --state "youtube-downloader/data/<handle>.json"
  • <handle> is the Handle column value without the @ prefix.
  • <Channel Name> is from the Channel Name column in Youtube Channels.md.
  • {x} defaults to 7 if not specified.
  1. Parse the JSON output from each scan.
  2. Report new videos found per channel:
Scan Results
============
All-In Podcast: 3 new videos found (N scanned)
  - Episode Title 1
  - Episode Title 2
  - Episode Title 3

Workflow: check -new

Flat scan without date filter (fast, IDs + titles only, default limit 21):

cd "C:/Users/10444/.claude/skills" && python _shared/run.py youtube_client.py scan --channel-url "<url>" --channel-name "<Channel Name>" --limit 21 --state "youtube-downloader/data/<handle>.json"

Workflow: download / download -{x}d

Default: download = download -7d (scan last 7 days, then download pending).

  1. Run check -{x}d workflow first (scan + update state with date filter; default 7 days).
  2. For each channel with pending episodes, run in background (timeout 600000):
cd "C:/Users/10444/.claude/skills" && python _shared/run.py youtube_client.py batch-download --state "youtube-downloader/data/<handle>.json" --output-dir "C:/Users/10444/Desktop/Claude Download/Info-Lake/Youtube-transcript" --filter pending --delay 3 --jitter 1.5
  1. Stream progress via TaskOutput (check every 15-20 seconds with block=false).
  2. When complete, report download summary per channel:
Download Summary
================
All-In Podcast: 3 downloaded, 0 failed, 1 unavailable

Workflow: download -new

Download pending transcripts without scanning first (default max 21 per channel):

  1. Read Start/Youtube Channels.md, parse table.
  2. For each channel with pending episodes, run in background (timeout 600000):
cd "C:/Users/10444/.claude/skills" && python _shared/run.py youtube_client.py batch-download --state "youtube-downloader/data/<handle>.json" --output-dir "C:/Users/10444/Desktop/Claude Download/Info-Lake/Youtube-transcript" --filter pending --max 21 --delay 3 --jitter 1.5
  1. Stream progress and report summary.

Channel Config Format

Start/Youtube Channels.md contains:

| Handle | Channel Name | URL | Default Language |
|--------|-------------|-----|-----------------|
| @allin | All-In Podcast | https://www.youtube.com/@allin/videos | en |

Parse this table to get channel entries. Skip header and separator rows.

State File

Per-channel JSON at youtube-downloader/data/{handle}.json (e.g., allin.json):

  • handle: @allin
  • channel_name: display name
  • last_scan: ISO timestamp
  • episodes: dict keyed by video ID with status, title, dates, file path, error info

Output Format

Transcripts are saved to Info-Lake/Youtube-transcript/{Channel Name}/{YYYY-MM-DD} - {Title}.md with YAML frontmatter including processed: false and empty tickers/topics for /pre-processing pipeline integration.

Important Notes

  • Always read Start/Youtube Channels.md fresh at invocation start.
  • State is saved after each individual download (crash-safe).
  • Anti-blocking: 3s + random jitter between downloads, exponential backoff on failure.
  • unavailable status (TranscriptsDisabled, VideoUnavailable) is permanent — never retried.
  • No API keys needed — both yt-dlp and youtube-transcript-api work without authentication.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.