Torrere docs · Getting a recipe in Torrere

Torrere read 86 of 100 recipe sites

Checked against the app on .

On September 22, 2026, Torrere read 86 of 100 recipe sites straight from their own pages, asking each for one recipe page from Torrere's own server. 14 of the 100 sites answered with a browser check or a refusal instead of the page. The table below names every site and what it did. Torrere runs the same test every month, and this page says when it last ran.

Torrere asked each site for one recipe page

Torrere's reader asked each site for one recipe page, once, from Torrere's server, with a two-second pause between sites. It named itself as Torrere/1.0 (+https://torrere.com/reader) and read each site's robots.txt first, the same as when you paste a link into Torrere. A site counts as read when it served the recipe page: Torrere takes the recipe from the page's own recipe data, or from the page's text when the site publishes no recipe data.

What each site did when Torrere asked

Of the 100 sites asked on September 22, 2026, Torrere read 86: 82 from the page's recipe data and 4 from the page's text. So 82 of the 86 recipe pages Torrere was served carried the recipe as schema.org recipe data, steps included. Of the others, 13 answered with a browser check and 1 refused the first request as too many.

Site What happened when Torrere asked
101 Cookbooks Read, from the page's recipe data
Alexandra's Kitchen Read, from the page's recipe data
Allrecipes Browser check
Ambitious Kitchen Read, from the page's recipe data
America's Test Kitchen Read, from the page's recipe data
Averie Cooks Read, from the page's recipe data
Baker Bettie Read, from the page's recipe data
Baking a Moment Read, from the page's recipe data
BBC Food Read, from the page's recipe data
BBC Good Food Read, from the page's recipe data
Belly Full Read, from the page's recipe data
Better Homes & Gardens Browser check
Betty Crocker Read, from the page's recipe data
Bon Appétit Read, from the page's recipe data
Brown Eyed Baker Browser check
Budget Bytes Read, from the page's recipe data
Cafe Delites Read, from the page's recipe data
Carlsbad Cravings Read, from the page's recipe data
Chef Savvy Read, from the page's recipe data
Chelsea's Messy Apron Read, from the page's recipe data
Cookie and Kate Read, from the page's recipe data
Cookies and Cups Read, from the page's recipe data
Cooking Classy Read, from the page's recipe data
Country Living Read, from the page's recipe data
Damn Delicious Read, from the page's recipe data
David Lebovitz Read, from the page's recipe data
delicious. magazine Read, from the page's recipe data
Delish Read, from the page's recipe data
Dinner at the Zoo Read, from the page's recipe data
Dinner, then Dessert Read, from the page's recipe data
Downshiftology Read, from the page's recipe data
EatingWell Browser check
Epicurious Read, from the page's recipe data
Feasting at Home Read, from the page's recipe data
Food & Wine Browser check
Food Network Read, from the page's recipe data
Food.com Read, from the page's recipe data
Food52 Refused as too many requests
Gimme Some Oven Read, from the page's recipe data
Good Housekeeping Read, from the page's recipe data
Half Baked Harvest Read, from the page's recipe data
Handle the Heat Read, from the page's recipe data
Hebbar's Kitchen Read, from the page's recipe data
Inspired Taste Read, from the page's text
Isabel Eats Read, from the page's recipe data
Jamie Oliver Read, from the page's recipe data
Jo Cooks Read, from the page's recipe data
Joy of Baking Read, from the page's text
Just One Cookbook Read, from the page's recipe data
King Arthur Baking Read, from the page's recipe data
Kitchen Sanctuary Browser check
Kristine's Kitchen Read, from the page's recipe data
Le Crème de la Crumb Read, from the page's recipe data
Lil' Luna Read, from the page's recipe data
Live Well Bake Often Read, from the page's recipe data
Love & Lemons Read, from the page's recipe data
Love and Olive Oil Read, from the page's recipe data
Maangchi Browser check
Martha Stewart Browser check
Mel's Kitchen Cafe Read, from the page's recipe data
Mexico in My Kitchen Read, from the page's recipe data
Milk Street Read, from the page's recipe data
Minimalist Baker Read, from the page's recipe data
Natasha's Kitchen Read, from the page's recipe data
Nigella Lawson Read, from the page's recipe data
Nora Cooks Read, from the page's recipe data
NYT Cooking Read, from the page's recipe data
olive Read, from the page's recipe data
Once Upon a Chef Read, from the page's recipe data
Pati Jinich Read, from the page's recipe data
Pillsbury Read, from the page's recipe data
Pinch of Yum Read, from the page's recipe data
Preppy Kitchen Read, from the page's recipe data
Rainbow Plant Life Read, from the page's recipe data
RecipeTin Eats Read, from the page's recipe data
Sally's Baking Addiction Read, from the page's recipe data
Salt & Lavender Read, from the page's recipe data
Saveur Read, from the page's text
Serious Eats Browser check
Simply Recipes Browser check
Skinnytaste Read, from the page's recipe data
Smitten Kitchen Read, from the page's text
Southern Living Browser check
Spend With Pennies Read, from the page's recipe data
Sugar Spun Run Read, from the page's recipe data
Swasthi's Recipes Read, from the page's recipe data
Taste of Home Read, from the page's recipe data
Tastes Better From Scratch Read, from the page's recipe data
Tasty Read, from the page's recipe data
The Kitchn Browser check
The Mediterranean Dish Read, from the page's recipe data
The Pioneer Woman Read, from the page's recipe data
The Recipe Critic Read, from the page's recipe data
The Spruce Eats Browser check
The Stay at Home Chef Read, from the page's recipe data
The Woks of Life Read, from the page's recipe data
Two Peas & Their Pod Read, from the page's recipe data
Vegan Richa Read, from the page's recipe data
Well Plated Read, from the page's recipe data
Wholesome Yum Read, from the page's recipe data

Torrere checked what each site's robots.txt tells other crawlers

A site's robots.txt says which crawlers may read which of its pages. Torrere reads it before every page it reads, and for this test it also checked what each site's file says to crawlers from OpenAI, Anthropic, Google, Perplexity, Common Crawl and ByteDance about the one recipe page Torrere tested there. A site counts when its file tells that crawler not to read that page.

On September 24, 2026, Torrere checked the robots.txt of 99 of the 100 sites against the recipe page it tested on each. 2 of them have no robots.txt, which leaves every page open to every crawler. 46 of the 99 name at least one of these crawlers in their robots.txt.

Whose crawler, and what it does Sites telling it not to read the recipe page
OpenAI, collecting pages to train its models 29 of 99
Anthropic, collecting pages to train its models 23 of 99
Google, deciding whether pages Google crawls may train its models 27 of 99
Common Crawl, collecting pages for its open archive of the web 37 of 99
ByteDance's crawler 23 of 99
OpenAI, finding pages to show in its search results 4 of 99
Anthropic, finding pages to show in its search results 13 of 99
Perplexity, finding pages to show in its search results 19 of 99
OpenAI, fetching a page a person asked about 9 of 99
Anthropic, fetching a page a person asked about 13 of 99
Perplexity, fetching a page a person asked about 14 of 99

Torrere's check reads the file and nothing else. A robots.txt line is a request, not a lock, and a site can also turn a crawler away at its server, which the file doesn't show. OpenAI and Perplexity say the fetchers they send when a person asks about a page may not follow robots.txt. The data file below names each crawler the way robots.txt does.

Torrere reads browser-checked sites from your phone

A site that answered with a browser check can still be read in the Torrere apps for iPhone and Android, which open the page in the phone's own browser view. When a site won't answer explains how Torrere does that. A site that refused Torrere, or asked it not to read its pages, isn't read from anywhere.

Torrere's result for a site can change

Sites change what they allow, sometimes from one week to the next, so a result here is what a site did on the day of the test rather than a promise. Torrere asks the same page on each site every month, and picks a new page from the same site only when the old one has gone. Torrere reads a recipe from a link covers what Torrere does with a page it can read.

Torrere shares the results as a data file

Every site, the page Torrere tested on it and what the site answered are in recipe-sites.csv, in Torrere's public repository. The file is under the CC BY 4.0 license, so anyone can use it as long as they say it came from Torrere.

Open Torrere