Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.lumilexicon.com/llms.txt

Use this file to discover all available pages before exploring further.

Quick Snap is the recommended intake method. Point your camera at a book and AI identifies it from the image — no barcode needed.

​
What to photograph

For the best results, photograph one of these (in order of reliability):
  1. Copyright page — Has publisher, date, edition statement, ISBN (if post-1970). Most reliable for identification.
  2. Title page — Title, author, publisher clearly printed. Great for pre-ISBN books.
  3. Front cover — Works when title and author are visible. Less reliable for common titles.
  4. Spine — Works in a pinch but has the least information.
For pre-1971 books without ISBNs, always photograph the copyright page and title page. These pages contain the edition statements and printing indicators that determine value.

​
How it works

  1. The photo is sent to the Quick Snap AI agent (Claude Haiku — fast and cheap)
  2. AI reads the image one digit at a time for ISBNs, and extracts title, author, publisher, and date
  3. If an ISBN is found, it’s cross-referenced with Open Library and Google Books for additional metadata
  4. The result is presented for your review — edit anything that looks wrong before saving

​
Tips

  • Steady, well-lit photos work best. Avoid glare on glossy covers.
  • If Quick Snap gets the title wrong, you can edit it before saving — or ask Lexi to fix it later.
  • Quick Snap costs 1 credit per photo (uses Claude Haiku).