NYT Cooking Recipe Scraper

Enumerate the complete NYT Cooking recipe catalog (~25K recipes) from the official sitemap and extract structured recipe data from the public schema.org Recipe JSON-LD embedded in each page.

What it collects

Every record contains the following fields:

Field	Type	Description
recipe_id	string	Unique NYT Cooking recipe identifier
url	string	Canonical recipe URL
name	string	Recipe title
author	string	NYT Cooking contributor byline
description	string	Recipe description / headnote
recipe_yield	string	Serving size (e.g. "4 servings")
total_time	string	Total cooking time (e.g. "1 hr 30 min")
prep_time	string	Preparation time
cook_time	string	Active cooking time
recipe_category	string	Meal category (e.g. "Dinner, Main Course")
recipe_cuisine	string	Cuisine style (e.g. "Mediterranean Inspired")
recipe_ingredient	array	List of ingredient strings with quantities
recipe_instructions	array	Step-by-step instructions
nutrition	string	JSON-serialized nutrition facts (calories, fat, carbs, protein, sodium, etc.) from schema.org NutritionInformation. null for recipes without nutrition data.
aggregate_rating	number	Average user rating (1–5 scale)
rating_count	integer	Number of user ratings
keywords	array	Tags and keywords (ingredient highlights, technique, difficulty, etc.)
image_urls	array	Full-resolution image URLs
date_published	string	ISO 8601 publication date

Discovery

By default the actor walks the official NYT Cooking sitemap index (https://www.nytimes.com/sitemaps/new/cooking.xml.gz), which contains monthly sub-sitemaps covering the full recipe inventory. Only /recipes/ paths are collected — article and guide pages are excluded.

Inputs

Input	Type	Default	Description
maxItems	integer	10	Maximum number of recipes to collect. Set to 0 for no limit (full catalog run).
startUrls	array	—	Optional list of NYT Cooking URLs — individual recipe pages to scrape directly, or a category/topic page (e.g. a cuisine or collection) to collect every recipe listed there. Bypasses sitemap discovery.

Data source

All data is extracted from the schema.org/Recipe JSON-LD markup that NYT Cooking embeds in every public recipe page for SEO purposes. Recipe content — including ingredients, instructions, and metadata — is publicly available. The NYT Cooking paywall only gates account-specific features (recipe box, personal notes, collections) and does not restrict access to recipe markup.

Usage notes

• For a full catalog run (~25K recipes), use maxItems: 0 and allow sufficient run time.
• Nutrition data (nutrition field) is present on most recipes but absent on some recently published ones; the field is null in those cases.
• The sitemap updates frequently (new recipes appear within hours of publication). Re-running with maxItems: 0 against the latest sub-sitemaps will catch additions.

---

Further reading: ISBN Database Access and Other Open Reference Data in Bulk