Searching 'quick pasta' shouldn't mean scrolling past someone's holiday photos.

Recipe sites are among the web's most SEO-saturated corners. Finding which of them actually has a good match for "mediterranean lunch" means opening seven tabs or trusting one aggregator's sponsored ranking.

The Recipe URL Scraper queries multiple recipe sites in parallel and interleaves the matches:

{
  "keywords": ["pasta", "mediterranean"]
}

{
  "keyword": "pasta",
  "title": "One-Pot Tomato Basil Pasta",
  "url": "https://www.budgetbytes.com/one-pot-tomato-basil-pasta/",
  "source": "budgetbytes"
}

Built to chain

The output shape is deliberately identical to the input of Recipe Extractor. Run them back to back: keyword search finds the candidates, the extractor turns each URL into structured ingredients and steps. Two calls, one clean recipe dataset.

The honest limit: search coverage depends on each site's own internal search, and a few engines rate-limit aggressively from datacenter IPs — results vary by keyword popularity. Popular terms match richly; obscure ones may return few hits.