Leaving, or keeping a copy
Everything you put in, in one file.
A ZIP of every recipe, collection, book and photograph in your kitchen, in formats that need nothing from RecipeJam to open. It works whether or not you are still subscribed.
Getting it
-
1
Sign in and choose Download everything from the account menu.
-
2
Choose Prepare my file. It is built in the background; most kitchens take a minute or two.
-
3
Come back to the same page and choose Download the file.
- The page stays open after a subscription ends, along with billing and your profile. Nothing in the kitchen can be changed until a subscription starts again, but all of it can be downloaded.
- The kitchen's owner makes the export. Other members of the kitchen are told who to ask.
- The link works for 7 days, and only for the owner while signed in. After that the file is deleted from our storage; your kitchen is not.
- One file at a time, and a fresh one at most every 60 minutes. A finished file can be downloaded as often as you like while its link lasts.
What is in the ZIP
recipejam-kitchen-2026-11-15.zip
README.txt
manifest.json
recipes.json
collections.json
books.json
schema-org/sunday-lamb-k3x9q2m1pz.jsonld
images/recipes/4krp3y5x13.jpg
images/steps/…
images/collections/…
images/books/…
books/christmas-q8w2e5r7t1.epub
Every file is UTF-8. The JSON files refer to each other, and to the photographs and eBooks, by id and by path inside the archive, so nothing points back at RecipeJam.
manifest.json
What the export is. Read this first: format and version tell a program whether it knows how to read the rest.
| Field | Type | Notes |
|---|---|---|
format
|
string | Always recipejam-account-export. |
version
|
integer | Currently 1. It goes up only when a field changes meaning or is removed; new fields can appear without it changing, so ignore what you do not recognise. |
exportedAt
|
string | ISO 8601. |
account
|
object | The kitchen's name and when it was created. |
exportedBy
|
object | The name and email of the person who asked for the export. |
counts
|
object | recipes, collections, books, photographs and ebooks in this file. |
files
|
object | Where each part is. |
recipes.json
One object, { format, version, recipes: [...] }, holding every recipe in full.
| Field | Type | Notes |
|---|---|---|
id
|
string | The recipe's id — the same one the API uses. |
name, description
|
string | As written. |
locale
|
string|null | The language the recipe is written in, e.g. en_GB. |
tags
|
string[] | |
timings
|
object | serves, prep, cook and total, as whole numbers (minutes for the times). Null where there is no figure. |
metadata
|
object | Everything else about the recipe as stored: servings, prepTime and cookTime as written, and the keeping notes (makeAhead, storage, freezing, reheating, serveWith). |
ingredients
|
object[] | Each line as { name, parsed }: name is exactly what was typed; parsed is our reading of it (quantity, unit, ingredient) and is a convenience, not the record. |
equipment
|
object[] | Each as { name }. |
prep, method
|
step[] | In order. Each step has id, position, text, durationMinutes, durationSource, unattended and images. |
notes
|
object[] | Each as { content }. |
source
|
object | Where the recipe came from: url, name and author, where known. |
links
|
object | Other links, such as video. |
images
|
image[] | Photographs of the finished dish. |
coverImage
|
string|null | The id of the photograph used as the cover. |
sharing
|
object | isPublic, shareNotes and shareKeeping, as set in RecipeJam. |
author
|
string|null | Who in the kitchen wrote it. |
createdAt, updatedAt
|
string | ISO 8601. |
schemaOrg
|
string | The path of the same recipe as schema.org JSON-LD. |
collections.json
{ format, version, collections: [...] }.
| Field | Type | Notes |
|---|---|---|
id, name, description, locale, tags
|
As for a recipe. | |
recipes
|
string[] | Recipe ids, in the collection's order. Only your own recipes — see below. |
images, coverImage
|
As for a recipe. | |
sharing
|
object | isPublic. |
createdAt, updatedAt
|
string | ISO 8601. |
books.json
{ format, version, books: [...] }.
| Field | Type | Notes |
|---|---|---|
id, name, locale, tags
|
As for a recipe. | |
metadata
|
object | The book's credits: title, author, byline and the rest. |
pages
|
object[] | In order. Each has id, position, type (cover, title, toc, collection, recipe, dedication…), origin (generated or placed), recipe and collection ids where the page prints one, image, and metadata — the words on the page. |
images, coverImage
|
As for a recipe. | |
sharing
|
object | isPublic and storeUrl. |
compiledAt
|
string|null | When the eBook was last built. |
epub
|
string|null | The path of the EPUB inside the archive, or null if it has never been built or prints a recipe from another kitchen. |
Photographs
Every photograph entry, wherever it appears, has the same shape.
| Field | Type | Notes |
|---|---|---|
id
|
string | The photograph's id. |
path
|
string|null | Where the file is inside the archive, e.g. images/recipes/abc123.jpg. Null if the stored file could not be found. |
width, height
|
integer | Pixels. |
bytes
|
integer|null | Size of the file. |
originalName
|
string | The name of the file as it was uploaded. |
sourceUrl
|
string|null | Where it came from, for a photograph imported with a recipe. |
provider, credit, creditUrl
|
string|null | Who took it and where to credit them, for a stock or imported photograph. |
createdAt
|
string | ISO 8601. |
The file is the largest copy RecipeJam keeps: a JPEG of up to 2048 pixels on its longest side, with camera metadata such as location removed when it was uploaded. The original upload is not kept, so it cannot be exported.
schema.org recipes
Each recipe again, on its own, as schema.org/Recipe JSON-LD — the format recipe websites publish and recipe apps import.
It carries the name, description, ingredients as written, prep and method as steps, servings, times, tags, equipment, the source address and the cover photograph (as a path relative to the file). schema.org/Recipe is the structure recipe websites embed in their pages, and it is what recipe managers and search engines read when they import or index a recipe from the web — so anything built to read it can read these. Whether a particular app accepts a file directly, and how much of it it keeps, varies from app to app; recipes.json is the complete record.
What is not included
- Recipes that belong to another kitchen, even when one of your collections holds them — sharing a recipe with you is not giving it to you. A book that prints such a recipe has its pages listed but no eBook file.
- Anything in Recently deleted. Restore it first to include it.
- Other members' account details, passwords, tokens and connections.
There is no import back into RecipeJam yet. The format is written so that one is possible, by us or by anybody else.
Support
Get support
Stuck on a connection, an endpoint or an export? Tell us what you tried and what happened, and a person will reply.