# Sprouts Extraction Notes

- Live account entry discovery (2026-04-10): homepage header `Sign In / Register` routes through Sprouts SSO to `https://account.sprouts.com/account`.
- Treat `account.sprouts.com/account` as the current verified account entry surface; from there, continue into `Your Orders` / `Order History` rather than relying on the dead `www.sprouts.com/account/order-history` route.
- Watch for first-load zip code entry modal; complete before scraping.
- Orders page may lazy-load; use repeated bottom scroll + load-more until stable.
- Always expand all items before extraction.
- Preserve repeated adjustment labels as separate rows.
- Treat item-list truncation as blocker.
- 2026-04-11 autonomous-path progress: `scripts/sprouts_extract_cdp.js --mode live` now self-attaches to CDP (port `9333`), launches automation Chrome with merchant profile if endpoint is absent, captures before/after/account/orders screenshots, and writes a live source artifact + stage payload without manual browser attach.
- Current live blocker remains parser/routing: after credential submission, Sprouts can route to rewards/signup surfaces instead of stable order-history DOM, so autonomous extraction currently preserves artifacts and returns 0 orders rather than risking fabricated data.