tsazan

↗ HN profile [ 59.8 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. OP here. I took the unofficial IKEA US dataset (originally scraped by jeffreyszhou) and converted all 30,511 products into a flat, markdown-like protocol called CommerceTXT. The goal: See if a flatter structure is more…

  2. Problem: AI agents waste 8,500+ tokens scraping HTML to find a product price. They hallucinate inventory and merchants have no control. Solution: CommerceTXT - a read-only protocol that gives AI deterministic commerce…

  3. Hi HN, author here. I built CommerceTXT because I got tired of the fragility of extracting pricing and inventory data from HTML. AI agents currently waste ~8k tokens just to parse a product page, only to hallucinate the…