The problem
Most scraping and data extraction tools treat every fetch as a replacement. You get the current state of a page, and that's it. If the source changes, breaks, or removes something, you have no history, no diff, and no way to roll back.
I got tired of rebuilding that missing layer (a database, a diffing script, somewhere to store snapshots) every time a project needed to track a page over time. So I built it as the product instead of rebuilding it every time.
What Quorel does
Point it at any public URL and describe the fields you want in plain English (no CSS selectors, no XPath). Quorel extracts them and:
- Saves every refresh as its own immutable version. Nothing is overwritten.
- Lets you diff any two versions to see exactly what changed: added, removed, modified.
- Lets you roll back to any prior version in one click if a refresh comes back bad.
- Refreshes nightly by default, or on demand via a ping URL, webhook, or your own cron.
Because extraction is driven by the fields you describe rather than fixed selectors, most layout changes on the source site don't silently break your data. You see it show up in the diff instead.
MCP included
Every dataset ships with an MCP server on every plan, including free. An agent (Claude, or anything MCP-compatible) can query the data directly, or pull the raw rows, fix nulls/duplicates, and publish a new version itself.
Output and pricing
Export as JSON, CSV, JSONL, XML, Excel, or Parquet, with filtering, sorting, and pagination built in.
Pricing is flat, not usage-metered:
- Free: 1 dataset, 20 URLs, full version history, MCP access
- Pro ($19/mo): 5 datasets, 100 URLs, on-demand refresh, webhooks
Try it
quorel.vercel.app, no card required.
I'd love feedback, especially from anyone who's been burned by a scraper that silently broke, or wished they'd kept last week's data to compare against.
