メニュー
YouTube Data API v3 complete wrapper - search videos, get video/channel/playlist details, get comments, download subtitles, etc. Supports filtering and sorting by time/views/rating and more.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Capture a Mac app window as agent context. Use when the user asks to snap, inspect, read, capture, summarize, or work from the current Mac window, a visible app, a selected window, or an app screenshot. Strongly prefer for Appshot-like requests, "what is on this window", "use this app as context", and "grab the window text". SKIP for remote websites/files that can be fetched directly, for controlling the UI, clicking, typing, automating actions, or extracting hidden Google Docs/Sheets/Slides/Gmail canvas text beyond what macOS Accessibility exposes.
Adaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
Load when the user or an agent wants to use Grok/xAI locally with their own SuperGrok / X Premium account instead of a paid xAI API key — chat with Grok, generate images or videos with Grok Imagine, or search X/Twitter posts via Grok. Triggers on "用 Grok 聊天/问 Grok", "Grok 生成图片/视频", "grok imagine", "搜一下推特/X 上怎么说", "grok-cli ...", "log in to Grok with my X Premium", "Grok image/video generation", "search tweets with Grok". Uses browser OAuth (no API key needed); image/video generation is paid. Do NOT load for non-Grok LLMs or for the paid xAI API key SDK workflow.
Load when the user wants to view, sync, or analyze multi-agent activity across multiple devices — Claude Code / Codex / Cursor / Maestri / browser-use sessions captured to a shared GitHub-synced pool, OR wants a distilled per-project context document so a fresh agent in a different tool can pick up where another left off. Triggers on "agentlog X" commands, "what did I do today across all my agents", "show me my pool", "sync the pool", "拉一下另一台机器上的 agent 记录", "看看 vps 上跑的", "跨设备 agent log", "跨工具上下文", "switch from Codex to Claude Code", "AGENTS.md / CLAUDE.md", "项目状态文档", "agent context". Do NOT load for single-machine tweet material capture (use `seed`) or one-off project planning.
Use when the user wants a real data analysis on a dataset — finding what drives a metric, testing whether a difference is real, building a predictive model, reducing many variables to a few, or interpreting what numbers mean for the business. Trigger phrases include "analyze this data", "what drives X", "is this significant", "build a model to predict", "find the factors", "do a proper analysis", "what does this data mean". Skip when the user only wants a chart, only wants raw data pulled with no question, asks a closed factual question with no data, or just wants APA-formatted stats output.
Expert cooking assistant for Chinese and Western cuisine. Provides step-by-step recipes grounded in real data from scraped websites (xiachufang, allrecipes, BBC Good Food, etc.), free APIs (TheMealDB, Open Food Facts), and Kaggle datasets (Food.com, Epicurious). Use when user asks to cook a specific dish, wants suggestions based on ingredients they have, asks about technique/substitution/pairing, or wants nutrition info. Recipes are cached as Markdown files in data/recipes/ with YAML frontmatter — never re-fetch what's already there.
| name | youtube-data-api |
| description | YouTube Data API v3 complete wrapper - search videos, get video/channel/playlist details, get comments, download subtitles, etc. Supports filtering and sorting by time/views/rating and more. |
| version | 1.0.0 |
| user-invocable | true |
| allowed-tools | ["Bash","Read","Write","Edit","WebFetch","Task"] |
Complete wrapper based on official YouTube Data API v3 documentation.
# Option 1: Environment variable (recommended)
export YOUTUBE_API_KEY="YOUR_API_KEY_HERE"
# Option 2: Edit config file directly
# Edit ~/.claude/skills/youtube-data-api/scripts/config.py
| Operation Type | Quota Cost (units) |
|---|---|
| Read operations (list) | 1 |
| Search requests | 100 |
| Write operations (insert/update/delete) | 50 |
| Video upload | 1600 |
Default daily quota: 10,000 units
Quota cost: 100 units/request
# Basic search
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python tutorial" \
--max-results 25
# Filter by time (last week)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "machine learning" \
--published-after "7d" \
--type video
# Sort by view count
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "React hooks" \
--order viewCount \
--type video
# HD videos only
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "4K nature" \
--video-definition high \
--type video
# Videos with captions
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "TED talk" \
--video-caption closedCaption \
--type video
# Live streaming videos
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "gaming" \
--event-type live \
--type video
# Filter by region
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "news" \
--region-code US \
--relevance-language en
# Filter by video duration
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "documentary" \
--video-duration long \
--type video
| Parameter | Description | Values |
|---|---|---|
--query, -q | Search keywords | Supports boolean operators (python AND tutorial) |
--type | Resource type | video, channel, playlist |
--max-results | Results count | 1-50, default 25 |
--order | Sort order | relevance(default), date, rating, viewCount, title |
--published-after | Published after | 1d, 7d, 30d, 1y or ISO 8601 format |
--published-before | Published before | Same as above |
--region-code | Region code | ISO 3166-1 (e.g., US, CN, JP) |
--relevance-language | Language | ISO 639-1 (e.g., en, zh, ja) |
--video-duration | Video length | short(<4min), medium(4-20min), long(>20min) |
--video-definition | Definition | high(HD), standard(SD) |
--video-caption | Captions | closedCaption(has captions), none(no captions) |
--video-dimension | Dimension | 2d, 3d |
--event-type | Live status | live(streaming), completed(ended), upcoming(scheduled) |
--channel-id | Limit to channel | Channel ID |
--safe-search | Safe search | none, moderate, strict |
Quota cost: 1 unit/request
# Get single video details
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--id "dQw4w9WgXcQ" \
--parts snippet,statistics,contentDetails
# Get multiple videos (batch)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--id "dQw4w9WgXcQ,jNQXAC9IVRw,9bZkp7q19f0" \
--parts snippet,statistics
# Get trending videos
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--chart mostPopular \
--region-code US \
--max-results 50
# Get trending by category
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--chart mostPopular \
--video-category-id 10 \
--region-code JP
| Part | Contains |
|---|---|
snippet | Title, description, thumbnails, channel info, publish time, tags |
statistics | View count, like count, comment count |
contentDetails | Duration, definition, caption info, region restrictions |
status | Upload status, privacy, license |
player | Embedded player HTML |
topicDetails | Related topic IDs |
recordingDetails | Recording location, date |
liveStreamingDetails | Live info (start/end time, viewer count) |
| ID | Category | ID | Category |
|---|---|---|---|
| 1 | Film & Animation | 20 | Gaming |
| 2 | Autos & Vehicles | 22 | People & Blogs |
| 10 | Music | 23 | Comedy |
| 15 | Pets & Animals | 24 | Entertainment |
| 17 | Sports | 25 | News & Politics |
| 19 | Travel & Events | 26 | Howto & Style |
| 27 | Education | 28 | Science & Technology |
Quota cost: 1 unit/request
# Get by channel ID
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py channel \
--id "UC_x5XG1OV2P6uZZ5FSM9Ttw" \
--parts snippet,statistics,contentDetails
# Get by username
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py channel \
--for-username "GoogleDevelopers" \
--parts snippet,statistics
# Get by handle (@username)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py channel \
--for-handle "@Google" \
--parts snippet,statistics,brandingSettings
| Part | Contains |
|---|---|
snippet | Channel name, description, thumbnails, country |
statistics | Subscriber count, video count, total views |
contentDetails | Related playlists (uploads, likes, etc.) |
brandingSettings | Channel banner, keywords, default language |
topicDetails | Channel topics |
status | Channel status (long video upload permission, etc.) |
Quota cost: 1 unit/request
# Get playlist by ID
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py playlist \
--id "PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf" \
--parts snippet,contentDetails
# Get all playlists for a channel
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py playlist \
--channel-id "UC_x5XG1OV2P6uZZ5FSM9Ttw" \
--max-results 50
Quota cost: 1 unit/request
# Get all videos in a playlist
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py playlist-items \
--playlist-id "PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf" \
--max-results 50
# Paginated retrieval
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py playlist-items \
--playlist-id "PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf" \
--page-token "NEXT_PAGE_TOKEN"
Quota cost: 1 unit/request
# Get all comments for a video (top-level)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py comments \
--video-id "dQw4w9WgXcQ" \
--max-results 100 \
--order relevance
# Get comments for a channel
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py comments \
--channel-id "UC_x5XG1OV2P6uZZ5FSM9Ttw" \
--max-results 50
# Get replies to a comment
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py comment-replies \
--parent-id "COMMENT_ID" \
--max-results 50
| Sort | Description |
|---|---|
relevance | By relevance (default) |
time | By time (newest first) |
Quota cost: 1 unit/request (list), 50 units (download)
# List all caption tracks for a video
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py captions \
--video-id "dQw4w9WgXcQ" \
--list
# Download captions (requires OAuth)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py captions \
--video-id "dQw4w9WgXcQ" \
--download \
--caption-id "CAPTION_ID" \
--format srt
Note: Downloading captions requires video owner authorization (OAuth). For public videos, use yt-dlp to get auto-generated captions.
# Install yt-dlp
pip install yt-dlp
# List available subtitles
yt-dlp --list-subs "https://www.youtube.com/watch?v=VIDEO_ID"
# Download subtitles (skip video)
yt-dlp --write-sub --write-auto-sub --sub-lang en,zh-Hans --skip-download \
"https://www.youtube.com/watch?v=VIDEO_ID"
# Download and convert to SRT format
yt-dlp --write-sub --write-auto-sub --sub-lang en --sub-format srt --skip-download \
"https://www.youtube.com/watch?v=VIDEO_ID"
Quota cost: 1 unit/request
# Get video categories for a region
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py categories \
--region-code US
# Get categories for China
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py categories \
--region-code CN
Quota cost: 1 unit/request
# Get supported regions list
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py regions
# Get supported languages list
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py languages
# Search videos about "AI" from last 7 days, sorted by view count
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "AI artificial intelligence" \
--published-after "7d" \
--order viewCount \
--type video \
--max-results 50 \
--output ai_trending.json
# Get detailed statistics for these videos
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--ids-from ai_trending.json \
--parts snippet,statistics,contentDetails \
--output ai_trending_details.json
# 1. Get channel info
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py channel \
--for-handle "@GoogleDevelopers" \
--parts snippet,statistics,contentDetails \
--output channel_info.json
# 2. Get uploads playlist ID (from contentDetails.relatedPlaylists.uploads)
# 3. Get all uploaded videos
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py playlist-items \
--playlist-id "UPLOADS_PLAYLIST_ID" \
--max-results 50 \
--all-pages \
--output channel_videos.json
# 1. Search videos
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "TED talk motivation" \
--video-caption closedCaption \
--type video \
--max-results 10 \
--output ted_videos.json
# 2. Download subtitles using yt-dlp
for video_id in $(jq -r '.items[].id.videoId' ted_videos.json); do
yt-dlp --write-sub --write-auto-sub --sub-lang en --skip-download \
"https://www.youtube.com/watch?v=$video_id" \
-o "subtitles/%(id)s.%(ext)s"
done
# Get all comments for a video
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py comments \
--video-id "VIDEO_ID" \
--max-results 100 \
--all-pages \
--output comments.json
# Comment data includes:
# - Author name
# - Comment text
# - Like count
# - Publish time
# - Reply count
All commands default to JSON output, save to file with --output parameter.
# Output to file
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python" \
--output results.json
# CSV format (tabular data)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python" \
--format csv \
--output results.csv
# Simple output (titles and IDs only)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python" \
--format simple
For large datasets, API uses pageToken for pagination:
# Auto-fetch all pages
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python" \
--all-pages \
--max-total 500 \
--output all_results.json
# Manual pagination
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py search \
--query "Python" \
--page-token "NEXT_PAGE_TOKEN"
| Code | Description | Solution |
|---|---|---|
| 400 | Bad request | Check parameter format |
| 401 | Unauthorized | Check API Key |
| 403 | Quota exceeded or insufficient permissions | Check quota or request increase |
| 404 | Resource not found | Verify ID is correct |
| 429 | Too many requests | Reduce request frequency |
fields parameter to request only needed fields# Only get needed fields (reduces response size)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py video \
--id "VIDEO_ID" \
--fields "items(id,snippet(title,channelTitle),statistics(viewCount))"
The following operations require user authorization (API Key alone is not sufficient):
For these operations, configure OAuth 2.0 client:
# Setup OAuth (first time use)
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py auth \
--client-secrets client_secrets.json
# Install Python dependencies
pip install google-api-python-client google-auth-oauthlib
# Optional: Install yt-dlp (for subtitle downloading)
pip install yt-dlp
# Check if API key is valid
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py test
# Check quota usage
python ~/.claude/skills/youtube-data-api/scripts/youtube_api.py quota