Import articles from URLs

How to get there: Go to your Knowledge Base board → open the actions menu → click Import from URL.

Migrate an existing help site into ProductLift by pasting a list of URLs (or a sitemap). Each page is scraped, cleaned with AI, and added as a knowledge base article. Images from the source page are automatically downloaded and rehosted on your storage.

When to use this

  • You are moving off an older help-desk tool and want to bring your existing articles into ProductLift.
  • A vendor gave you a public help site and you want a private, editable copy.
  • You want a starting point rather than writing every article from scratch.

Input formats

The URLs or sitemap XML field accepts either:

  • A list of URLs, one per line. Empty lines are ignored.
  • A sitemap XML. Paste the raw <urlset>...</urlset> and every <loc> entry is picked up.

Only http:// and https:// URLs are used. Duplicates are removed automatically.

Per-plan limits

  • Trial: up to 20 URL-imported articles in total
  • Paid plans: up to 100 URL-imported articles in total

These are lifetime caps, not per-batch limits. Every successful scrape counts, including re-imports of the same URL (re-running the import re-scrapes the page and re-runs the AI clean pass, so it costs the same as a fresh import). Need more? Reply to your last invoice email or contact support and we can raise the cap for your account.

What happens per URL

  1. Scrape — the page is fetched using the Firecrawl scraper, keeping only the main content.
  2. Clean — a Gemini Flash Lite pass strips site chrome (nav, breadcrumbs, footer, video-player UI, "print this topic" boilerplate, etc.) without rewording the article.
  3. Rehost images — every image referenced in the article is downloaded and uploaded to your storage. The original external URLs are replaced with your storage URLs, so you can turn off the old site later without breaking anything.
  4. Publish — the article is inserted into a section called Imported from URL. Re-running with the same URL updates the existing article instead of creating a duplicate (and still counts against your cap because it re-runs the full scrape + AI pipeline).

Reorganizing after import

Imported articles all land in one section. Use the Manage Structure action on the same actions menu to rename articles, move them into other sections, or create a new topic hierarchy that matches your KB.

Editor formats

Articles are stored in your portal's editor format:

  • WYSIWYG portals: the cleaned markdown is converted to HTML before saving, so it renders and edits exactly like any other WYSIWYG article.
  • Markdown portals: the cleaned markdown is stored as-is.

Progress and errors

The import runs in the background. You can close the window and come back later — the summary at the end shows how many articles were created, updated, and (if any) failed, with the source URL and reason for each failure.