Recipe Scraper

Extract recipes with ingredients, instructions, nutrition info, ratings, and cooking times from 100+ cooking websites using Schema.org structured data.

Why use this actor?

Need recipe data at scale? This scraper works with any website that uses Schema.org Recipe markup -- which includes virtually every major cooking site. Get clean, structured recipe data for meal planning apps, nutrition analysis, recipe aggregators, or content research. No site-specific configuration needed.

Supported sites

Works out of the box with any site using Schema.org Recipe markup, including:

• allrecipes.com
• bbcgoodfood.com
• food.com
• simplyrecipes.com
• delish.com
• epicurious.com
• bonappetit.com
• tasty.co
• foodnetwork.com
• seriouseats.com
• smittenkitchen.com
• budgetbytes.com
• Any WordPress blog with recipe plugins (WP Recipe Maker, Tasty Recipes, etc.)

Input

Field	Type	Required	Description
startUrls	Array	No	Recipe page URLs or search result pages to scrape
searchQuery	String	No	Search for recipes by keyword (e.g., "pasta carbonara")
maxResults	Integer	No	Maximum recipes to return (default: 100)
proxyConfiguration	Object	No	Proxy settings (recommended for protected sites like allrecipes.com)

Output

Each result contains:

Field	Type	Example
name	String	"Chicken Piccata"
description	String	"A quick and easy Italian classic..."
author	String	"Chef John"
prepTime	String	"15min"
cookTime	String	"20min"
totalTime	String	"35min"
servings	Number	4
calories	Number	350
ingredients	Array	["4 chicken breasts", "1/2 cup flour", ...]
instructions	Array	["Pound chicken to even thickness...", ...]
cuisine	String	"Italian"
category	String	"Main Course"
rating	Number	4.8
reviewCount	Number	342
keywords	Array	["chicken", "piccata", "lemon"]
nutrition	Object	{"Calories": "350 calories", "Protein": "35g", ...}
imageUrl	String	Full image URL
url	String	Source recipe URL
scrapedAt	String	ISO timestamp

Example output

{
    "name": "Classic Banana Bread",
    "url": "https://www.allrecipes.com/recipe/20144/banana-banana-bread/",
    "scrapedAt": "2026-04-03T12:00:00.000Z",
    "description": "This is the best banana bread recipe. It's easy to make and turns out perfectly moist every time.",
    "author": "Shelley Albeluhn",
    "imageUrl": "https://www.allrecipes.com/thmb/...",
    "prepTime": "15min",
    "cookTime": "1h 5min",
    "totalTime": "1h 20min",
    "servings": 12,
    "calories": 232,
    "ingredients": [
        "2 cups all-purpose flour",
        "1 teaspoon baking soda",
        "1/4 teaspoon salt",
        "3 bananas, mashed",
        "3/4 cup white sugar",
        "1 egg, beaten",
        "1/3 cup melted butter"
    ],
    "instructions": [
        "Preheat oven to 350 degrees F (175 degrees C). Lightly grease a 9x5-inch loaf pan.",
        "Combine flour, baking soda, and salt in a large bowl.",
        "Mix bananas, sugar, egg, and melted butter in a separate bowl.",
        "Stir banana mixture into flour mixture until just combined.",
        "Pour batter into prepared loaf pan.",
        "Bake for 60 to 65 minutes, until a toothpick inserted into center comes out clean."
    ],
    "cuisine": "American",
    "category": "Bread",
    "rating": 4.7,
    "reviewCount": 12847,
    "keywords": ["banana bread", "quick bread", "baking"],
    "nutrition": {
        "Calories": "232 calories",
        "Fat": "7g",
        "Carbohydrates": "40g",
        "Protein": "3g"
    }
}

How it works

1. JSON-LD extraction (primary) -- Parses <script type="application/ld+json"> tags for Schema.org Recipe markup. This is the most reliable method and works on 95%+ of recipe sites.
2. HTML fallback -- When JSON-LD is not available, falls back to parsing common HTML selectors and meta tags.
3. Search support -- Automatically discovers recipe links from search result pages and follows them.

Cost estimate

This actor uses approximately 0.005 compute units per recipe (Cheerio-based, no browser needed). At standard Apify pricing, that's roughly $0.25 per 1,000 recipes.

Limitations

• Some sites (allrecipes.com, bbcgoodfood.com) have anti-bot protection -- use proxy configuration for best results
• HTML fallback produces less complete data than JSON-LD extraction
• Search pagination support varies by site
• Data is scraped from public websites and may change without notice