agentleFS
valtterimelkkovaltterimelkko/agent-workflow-skills/skills/youtube-search/SKILL.md

youtube-search

Skill2 starsChanged 46 days agoMITView on GitHub

Search YouTube videos by query or within specific channels. Supports date filtering, sort ordering, and duration filters to exclude YouTube Shorts. Returns LLM-optimized JSON with video details.

  • Reads credentials

13 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use youtube-transcript.

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.
  • dandacompanySkill

    youtube-transcript

    Extract YouTube video transcripts/subtitles using yt-dlp. Handles rate limiting, auto-generated captions, multiple languages, and VTT parsing.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search for YouTube videos using the YouTube Data API v3.

    line 8
  2. Distinctive:

    Search videos by topic or channel

    line 23
  3. Distinctive:

    Use channel_id, not handle: when possible (saves resolution cost)

    line 133
  4. Very specific:

    For bulk channel analysis: Prefer youtube_channel_videos.py over youtube_channel_search.py

    line 137
  5. Distinctive:

    If you exhaust today's quota:

    line 141
  6. Distinctive:

    Option 1: Wait 24 hours for quota reset

    line 142
  7. Distinctive:

    Option 2: Create new Google Cloud project with fresh quota

    line 143

“Common” is measured: kept by at least 6 in 10 of the 12 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

Search YouTube Videos

Search for YouTube videos using the YouTube Data API v3. Find videos by query across all of YouTube or within specific channels.

Prerequisites

YouTube API Key available in the environment or a shell startup file such as ~/.bashrc:

export YOUTUBE_API_KEY="your-api-key-here"

Get one from Google Cloud Console: Create project → Enable YouTube Data API v3 → Create API Key.

Daily quota: 10,000 units (each search ≈ 100 units, channel resolution ≈ 1 unit, playlist fetch ≈ 1 unit).

When to Use

  • Search videos by topic or channel
  • Filter by date, duration, sort order
  • Get up to 50 results per query

Quick Start

Basic search:

python3 ./scripts/youtube_search.py --query "AI tutorials" --max-results 10

Channel-specific (if you already have the channel ID):

python3 ./scripts/youtube_search.py --query "Python" --channel-id "UC123..." --max-results 20

Parameters

ParameterRequiredDefaultNotes
--query or -qYes-Search term
--channel-id or -cNo-If provided, search only in this channel
--max-results or -mNo10Max 50 (preserves quota and LLM context)
--order or -oNorelevancedate, rating, relevance, viewCount
--published-afterNo-ISO 8601 format (e.g., "2025-01-01T00:00:00Z")
--published-beforeNo-ISO 8601 format
--video-duration or -dNoanyany, short (<4min), medium (4-20min), long (>20min)

Helper Scripts (API Quota Warning)

⚠️ Each helper script uses API quota. Plan carefully before running multiple times.

1. youtube_channel_id.py - Resolve @handle to Channel ID

Cost: 1 unit per resolution

python3 ./scripts/youtube_channel_id.py --handle "@MetalSole"

2. youtube_channel_videos.py - List Channel Videos

Cost: 1-2 units for channel lookup + 1 unit per ~50 videos fetched

# By handle
python3 ./scripts/youtube_channel_videos.py --handle "@MetalSole" --max-results 50

# By channel ID (saves quota - skips handle resolution)
python3 ./scripts/youtube_channel_videos.py --channel-id "UC123..." --max-results 50

# By playlist ID (cheapest - fastest method)
python3 ./scripts/youtube_channel_videos.py --playlist-id "UU123..." --max-results 50

3. youtube_channel_search.py - Search Within Channel

Cost: 1-2 units (channel lookup) + 100 units (search query)

python3 ./scripts/youtube_channel_search.py --channel-handle "@MetalSole" --query "AI"

Quota Estimation

OperationCostPurpose
Basic search with query~100Find videos by topic
Resolve handle (@MetalSole)~1Get channel ID from handle
Fetch 50 channel videos~1List all uploads from channel
Search within channel~100Find topic within specific channel

Example cost: Searching 10 channels for videos = ~10 units. Then searching within each channel = ~1000 units total. Plan accordingly!

Output Format

Returns JSON with video metadata:

{
  "success": true,
  "videos": [
    {
      "title": "Video Title",
      "url": "https://www.youtube.com/watch?v=VIDEO_ID",
      "channel_title": "Channel Name",
      "published_at": "2025-01-15T10:00:00Z",
      "description": "..."
    }
  ]
}

Examples

# Latest AI videos
python3 ./scripts/youtube_search.py --query "AI" --order "date" --max-results 10

# Longer-form content only (no shorts)
python3 ./scripts/youtube_search.py --query "tutorial" --video-duration "medium" --max-results 15

# Last 30 days
python3 ./scripts/youtube_search.py --query "news" --published-after "2025-11-23T00:00:00Z"

Troubleshooting

ProblemSolution
API key not foundMake available in your environment or shell startup file: export YOUTUBE_API_KEY="key-here"
403 ForbiddenKey lacks permissions. Enable YouTube Data API v3 in Google Cloud Console
Quota exceededUsed 10,000 units today. Create new Google Cloud project or wait 24hrs
No resultsTry broader query, remove date filters

Tips for Efficient Quota Use

  1. Use channel_id, not handle when possible (saves resolution cost)
  2. Use playlist_id for channel videos (cheapest - ~1 unit instead of 100+)
  3. Batch operations: Plan all queries before executing to avoid re-runs
  4. Cache results: Save output to file to avoid re-querying same channels
  5. For bulk channel analysis: Prefer youtube_channel_videos.py over youtube_channel_search.py

When Quota Runs Out

If you exhaust today's quota:

  • Option 1: Wait 24 hours for quota reset
  • Option 2: Create new Google Cloud project with fresh quota
  • Option 3: Request quota increase (takes 24-48 hours)
  • Option 4: Switch to web-based channel browsing (manual, no quota cost)

Security & Limitations

  • API key = read-only access to public data
  • Cannot access: private videos, watch history, subscriptions
  • Max 50 results per query (for context window)
  • View counts/likes not included

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.