youtube-content

YouTube Content SOP

Search YouTube, fetch video and channel metadata, extract transcripts, and paginate through results via the YouTube Data API v3.

When to Use

• User wants to search YouTube for videos or channels
• User wants to get video details (duration, view count, tags, description)
• User wants channel statistics or upload history
• User wants to extract a video transcript or subtitles
• User wants to paginate through a large result set

---

Part 1 — Setup

1. Get an API Key

1. Go to https://console.cloud.google.com
2. Create a project (or select an existing one)
3. Enable YouTube Data API v3 under APIs & Services → Library
4. Go to APIs & Services → Credentials → Create Credentials → API key
5. (Recommended) Restrict the key to YouTube Data API v3 and your IP

2. Environment Variable

export YOUTUBE_API_KEY="AIzaSy..."

Or in ~/.cowrangler/credentials.env:

YOUTUBE_API_KEY=AIzaSy...

3. Shell Helper

YOUTUBE_API_KEY="${YOUTUBE_API_KEY:-$(grep '^YOUTUBE_API_KEY=' ~/.cowrangler/credentials.env 2>/dev/null | cut -d= -f2 | tr -d '\n\r')}"
YT_BASE="https://www.googleapis.com/youtube/v3"

yt_get() {
  # $1 = endpoint path + params (already URL-encoded)
  curl -s "${YT_BASE}/$1&key=${YOUTUBE_API_KEY}"
}

---

Part 2 — Search

Search Videos

QUERY=$(python3 -c "import urllib.parse; print(urllib.parse.quote('python async tutorial'))")

yt_get "search?part=snippet&type=video&q=${QUERY}&maxResults=10&order=relevance" \
| python3 -c "
import sys, json
data = json.load(sys.stdin)
for item in data.get('items', []):
    vid_id = item['id']['videoId']
    title  = item['snippet']['title']
    channel = item['snippet']['channelTitle']
    published = item['snippet']['publishedAt'][:10]
    print(f'{vid_id}  [{published}]  {channel:30s}  {title[:60]}')
print()
print('nextPageToken:', data.get('nextPageToken', '(none)'))
"

Search Channels

QUERY=$(python3 -c "import urllib.parse; print(urllib.parse.quote('machine learning'))")

yt_get "search?part=snippet&type=channel&q=${QUERY}&maxResults=5" \
| python3 -c "
import sys, json
data = json.load(sys.stdin)
for item in data.get('items', []):
    ch_id   = item['id']['channelId']
    title   = item['snippet']['channelTitle']
    desc    = item['snippet']['description'][:80]
    print(f'{ch_id}  {title:30s}  {desc}')
"