public-data-without-api-keyslisted

# Getting real platform data without keys The rule this serves: **never fabricate content for a real person's site.** If the video titles are not real, don't ship video titles. These are the methods that made real data available without an API key. Public, published information only. Nothing behind a login, no rate-limit abuse, no bot-check evasion. --- ## YouTube — video IDs and titles, no key The DOM is lazy-loaded, so querying anchors returns nothing. The data is in the page's embedded state. **1. Get the IDs from `ytInitialData`:** ```js // on https://www.youtube.com/@<handle>/videos JSON.stringify(window.ytInitialData).match(/"videoId":"([\w-]{11})"/g) ``` Do **not** try to walk the JSON for paired title+id — the modern shape hides the pairing, and attempts to reconstruct it produce mismatched titles, which is worse than no titles. **2. Get title and ownership per ID from oEmbed** (public, no key, no quota): ```bash curl -s "https://www.youtube.com/oembed?url=https://www.youtube.com/watch?v=<ID>&format=json" # → {"title": "...", "author_name": "...", "thumbnail_url": "..."} ``` **3. Filter on `author_name`.** A channel page can surface videos the channel does not own. Matching the author name is what confirms authorship. **Thumbnails are permanent and safe to hardcode:** ``` https://i.ytimg.com/vi/<ID>/hqdefault.jpg ``` ## Facebook — post metadata via the crawler user-agent Facebook serves **full Open Graph metadata server-side to its own crawler UA**, wit