Skip to content
Corvid

YouTube

Looks up metadata for a YouTube video from a pasted URL (youtube.com/watch, youtu.be, /shorts, /embed, or /live links all work) — title, author/channel, description, duration, publish date, keywords, and thumbnails at every standard resolution.

Two tiers of data are available:

  • Keyless — always available, no setup required. Combines YouTube’s public oEmbed endpoint (title, author, thumbnail, embed HTML) with a best-effort scrape of the video page’s Open Graph/schema.org tags for fields oEmbed doesn’t expose (description, duration, publish date, keywords). Thumbnail URLs at every standard resolution are constructed directly, no extra request needed — higher resolutions (e.g. maxresdefault) may not exist for every video.
  • Extended stats — needs a free YouTube Data API v3 key (console.cloud.google.com), configured under Settings → API Keys like any other provider. Unlocks view/like/comment counts, tags, and category, on top of the keyless data above.

That same API key also unlocks the comments panel — there’s no keyless tier for comments. Load top comments (newest-first or top-first) with pagination, or search: YouTube’s API has no native comment-text search, so a search scans multiple pages server-side, filtering by author or text. A heavily-commented video’s search stops after a page/result cap rather than scanning forever — a “load more” button continues the scan from where it left off.

Results are ephemeral, like Domain Finder and Dork Runner — no history is persisted. Pasting a YouTube link into the command palette (/ or ⌘K/Ctrl+K) suggests this module directly.