Data Catalog

Everything this site knows about its own recipes, published as machine-readable JSON. These are not exports or copies. They are the exact files the recipe index, Dinner Rescue, Use What You’ve Got, and the related-recipe modules run on, so what you read here is what a visitor gets.

Each file carries a version string and, where it is derived from another file, a generatedFrom field recording where it came from. Start with the taxonomy: it defines the vocabulary every other dataset uses.

Datasets

Dataset	What it contains	Size
Recipe Index recipes-index.json	Every published recipe as a single JSON document. Each entry carries id, slug, canonical url, title, description, image, total minutes, difficulty, and its controlled taxonomy tags. This is the master record the rest of the catalog derives from.	1247 KB
Discovery Taxonomy taxonomy.json	The controlled vocabulary the entire site is tagged against, across eight axes: meal, mainIngredient, method, difficulty, occasion, cuisine, equipment and diet. Also defines alias mappings and the derived time-and-ease browse rules. Read this first — it explains the values used in every other dataset here.	3 KB
Recipe Recommendation Graph recommendation-graph.json	A deterministic graph over the recipe corpus. Every node records a signature ingredient and typed edges to other recipes: similarTo, usesSameIngredient, usesSameEquipment, pairsWith and contextLinks. The construction rules are recorded in the file’s own methodology field rather than left implicit.	1996 KB
Pantry Match Index pantry-index.json	Ingredient-to-recipe matching, built to answer "what can I cook with what I already have". Includes 180 ingredient suggestions and a list of assumed staples. Salt, black pepper, water and common cooking oils are excluded from match scoring; that rule is stated in the file’s scoringNote field.	1313 KB
Curated Collections collections.json	Collection definitions with their membership rules. Each records slug, title, standfirst, introduction, the rule that determines membership, a minimum size, status, and the resolved recipe slugs — so any collection can be reproduced rather than taken on trust.	78 KB
Search Synonym Groups search-synonyms.json	Synonym groups mapping colloquial ingredient and equipment names onto canonical taxonomy values — garbanzo to chickpea, crockpot to slow cooker, and so on. Small, but it is what makes plain-language queries resolve against the taxonomy.	1 KB
Seasonal Discovery Map seasonal-discovery.json	Month-to-theme mapping across seven seasonal windows, used to surface timely recipes. Each season lists its months, a heading, and the recipes it promotes.	8 KB

The taxonomy

Every recipe is tagged against eight axes. The same values drive the category hubs, the Recipe Index facets and the collection membership rules, so a tag means the same thing everywhere on the site.

• meal (9)
• mainIngredient (7)
• method (11)
• difficulty (3)
• occasion (7)
• cuisine (6)
• equipment (4)
• diet (7)

Alias mappings handle the ways people actually type things, and a derived browse layer turns cooking times into ranges such as 30-minute dinners and under 1 hour. Both are defined in taxonomy.json.

How this data is verified

Structured data is only useful if it agrees with the page a reader sees. On this site that is checked rather than assumed:

• All 506 recipes carry Recipe, BreadcrumbList and FAQPage JSON-LD, with every field Google lists as required or recommended populated.
• Visible FAQ text and FAQ structured data are compared question by question and answer by answer; they match word for word on every recipe.
• Prep, cook and total times are checked so that no recipe claims a total shorter than its own steps.
• Collection membership is generated from stated rules, not hand-picked, so it can be reproduced from the rule and the recipe index alone.

Agents and crawlers: a plain-text summary of this catalog and the rest of the site lives at /llms.txt.