Open Food Facts Scraper

Scrape the open Open Food Facts catalog — 3+ million grocery and packaged-food products worldwide, contributed by users and brands. Search by text or filter by barcode, brand, category, country, store (retailer), label, manufacturer, or product URL. Returns per-product nutrition facts (per 100 g), Nutri-Score, Eco-Score, NOVA processing classification, ingredients, allergens, packaging tags, and high-quality product images. HTTP-only via the public openfoodfacts.org JSON API. No auth, no cookies, no proxy required.

What this actor does

• Nine modes: search, byBarcodes, byCategory, byBrand, byCountry, byStore, byLabel, byManufacturer, byUrls
• Universal IDs: EAN-13 / EAN-8 / UPC-A barcodes
• Filters: country, language, NOVA group (multi-select), Nutri-Score (multi-select), Eco-Score (multi-select), containsLabels (organic / vegan / fair-trade / etc.), excludesAllergens (milk / gluten / peanuts / etc.), minProductCompleteness (0-100%), free-text containsKeyword
• Per-100 g nutriments: energy, fat, saturated fat, carbohydrates, sugars, added sugars, fiber, proteins, salt, sodium, cholesterol, iron, calcium, vitamins A/C/D
• Multi-region: the actor automatically routes country-filtered queries to the per-country API host (us.openfoodfacts.org, fr.openfoodfacts.org, etc.) for the freshest, most localized data
• Empty fields are always omitted — no null clutter

Output per product

• code — barcode (EAN-13 / EAN-8 / UPC-A)
• productName, primaryBrand, brands[], brandsTags[]
• quantity, packaging, packagingTags[]
• manufacturingPlaces
• stores[], storesTags[]
• countries[], countriesTags[]
• language — ISO 639-1
• categoryHierarchy[], categoriesTags[], mainCategory
• nutriments — per-100 g object: energyKcal, energyKj, fat, saturatedFat, carbohydrates, sugars, addedSugars, fiber, proteins, salt, sodium, transFat, cholesterol, iron, calcium, vitaminA, vitaminC, vitaminD, alcohol
• servingSize, servingQuantity
• nutriscore — A / B / C / D / E / unknown
• ecoscore — A / B / C / D / E / unknown
• novaGroup — 1 (unprocessed) → 4 (ultra-processed)
• additivesTags[], allergensTags[], labelsTags[], ingredientsTags[], tracesTags[]
• ingredientsText
• ingredientsAnalysis — palmOilStatus, veganStatus, vegetarianStatus
• imageUrls — front, ingredients, nutrition
• productCompleteness — 0.0–1.0
• createdAt, lastModifiedAt
• productUrl
• recordType: "product", scrapedAt

Input

Field	Type	Default	Description
mode	string	search	One of search, byBarcodes, byCategory, byBrand, byCountry, byStore, byLabel, byManufacturer, byUrls
searchQuery	string	chocolate	Free-text query — name, brand, or ingredient
barcodes	array	–	Barcodes (mode=byBarcodes) — EAN-13 / EAN-8 / UPC-A; dashes are stripped
categorySlug	string	–	Category slug (mode=byCategory) — pick from dropdown or enter custom
brandSlug	string	–	Brand slug (mode=byBrand) — e.g. nutella, coca-cola
storeSlug	string	–	Retailer slug (mode=byStore) — carrefour, walmart, aldi, etc.
labelSlug	string	–	Label slug (mode=byLabel) — organic, vegan, fair-trade, etc.
manufacturerSlug	string	–	Manufacturer / manufacturing-place slug (mode=byManufacturer)
productUrls	array	–	Direct product URLs (mode=byUrls)
country	string	any	Restrict results + drive country API host (United States, France, Germany, …)
language	string	any	Preferred display language (ISO 639-1)
novaGroup	array	–	Multi-select 1–4 (food-processing level)
nutriscore	array	–	Multi-select A / B / C / D / E / unknown
ecoscore	array	–	Multi-select A / B / C / D / E / unknown
containsLabels	array	–	Require selected labels (organic, vegan, gluten-free, …)
excludesAllergens	array	–	Drop products containing selected allergens (milk, gluten, peanuts, …)
minProductCompleteness	int	0	Drop products below this completeness % (0–100)
containsKeyword	string	–	Post-fetch substring filter on name / brand / ingredients
maxItems	int	50	Hard cap on emitted records (1–1000)

Example: search for chocolate, US market

{
  "mode": "search",
  "searchQuery": "chocolate",
  "country": "united-states",
  "maxItems": 15
}

Example: lookup by barcode

{
  "mode": "byBarcodes",
  "barcodes": ["3017620422003", "0037466016450", "5000159484695"]
}

Example: organic + vegan chocolate spreads with Nutri-Score A or B

{
  "mode": "byCategory",
  "categorySlug": "chocolate-spreads",
  "containsLabels": ["organic", "vegan"],
  "nutriscore": ["A", "B"]
}

Example: all Nutella products

{
  "mode": "byBrand",
  "brandSlug": "nutella"
}

Example: products sold at Carrefour, France

{
  "mode": "byStore",
  "storeSlug": "carrefour",
  "country": "france",
  "maxItems": 100
}

Example: gluten-free products

{
  "mode": "byLabel",
  "labelSlug": "gluten-free",
  "excludesAllergens": ["gluten", "milk"]
}

Example: products from a country, narrowed by category

{
  "mode": "byCountry",
  "country": "italy",
  "categorySlug": "pastas",
  "maxItems": 50
}

Example: lookup by product URL

{
  "mode": "byUrls",
  "productUrls": [
    "https://world.openfoodfacts.org/product/3017620422003/nutella"
  ]
}

Use cases

• CPG market research — track product launches, brand portfolios, packaging trends, NOVA-classification shifts across markets
• Retailer / pricing intelligence — discover the catalog footprint of major retailers (Walmart, Carrefour, Aldi, Lidl, Tesco) in each country
• Nutrition-app data ingestion — bulk-import barcode-keyed nutrition facts to power calorie counters, fitness trackers, and recipe scorers
• Diet & health research — segment products by Nutri-Score / Eco-Score / NOVA group; quantify additive prevalence by category
• Labeling compliance — find products by allergen or label (organic, vegan, gluten-free, fair-trade, kosher, halal)
• Sustainability research — Eco-Score and packaging tags surface recyclability, palm-oil status, and origin data
• E-commerce enrichment — fill missing brand / nutrition / ingredient gaps on your own catalog using barcode lookups
• Academic / public-health studies — bulk-export an entire category for longitudinal analysis

Data source

This actor reads data from Open Food Facts — a free, open, collaborative catalog of food products from around the world contributed by ~25,000 volunteers and brands under an Open Database License (ODbL). The OFF API is public and no API key is required. All requests are made with a polite User-Agent and a small inter-request delay.

FAQ

What's Open Food Facts? An open, crowd-sourced food product database with 3+ million products from 150+ countries. Like Wikipedia for groceries — anyone can add or edit a product, with photo evidence of the front, ingredients, and nutrition panels.

Is there a rate limit? No documented hard cap. The actor uses small inter-request delays (≈200 ms) and retries on the temporary 503 throttle that the CDN issues for very-large unbounded queries.

Why does mode=byCountry need a narrowing input on the world API? An unfiltered country-wide query against world.openfoodfacts.org would scan all 900,000+ US-tagged products (or 1.5M+ French products) — the CDN throttles those. The actor either routes to the per-country host (us.openfoodfacts.org, fr.openfoodfacts.org) or asks for a searchQuery / categorySlug to narrow the result.

What's the difference between Nutri-Score and Eco-Score? Nutri-Score (A=best, E=worst) is a French/European front-of-pack nutrition label scored against fat / saturated fat / sugar / salt vs protein / fiber / fruit-veg. Eco-Score combines life-cycle assessment, packaging, transport, and certifications into an A–E environmental rating.

What does NOVA group mean? A food-processing classification developed by the University of São Paulo. Group 1 = unprocessed (apples, rice). Group 2 = processed culinary ingredients (oils, salt). Group 3 = processed foods (canned beans). Group 4 = ultra-processed (sodas, packaged snacks, ready meals).

Why are some products missing nutrition data? Open Food Facts is crowd-sourced — completeness varies by product. Use the minProductCompleteness filter to require records that include the most-important fields.

Can I get historical prices? No, Open Food Facts does not store price history. It's a product catalog, not a marketplace tracker. For pricing data, combine these records with retailer-specific scrapers.

Why are barcodes left-padded with zeros? Open Food Facts canonicalizes barcodes as 13-digit strings (UPC-A is left-padded with a leading zero to align with EAN-13). The actor preserves the canonical form.