Ramen Lab — Privacy Policy

Effective date: August 13, 2026

1. Introduction & scope

Ramen Lab ("the app," "we," "us," "our") is an iOS app published by Ramen Lab Apps. This Privacy Policy explains what information the app does and does not handle, and the choices you have.

The app is local-first and anonymous by default: you can use all of it without creating an account, and the content you create lives on your device. There is one exception, and it happens only at your explicit request: when you use the 清書 (fair copy) organize feature, the draft you tap the seal on is sent to our server to be organized, as described in section 2. Creating an account is optional (section 2), and the name you give the app during setup never leaves your device (section 2).

We have intentionally built this app to collect as little as possible. We do not serve ads, we do not track you across apps or websites, and we do not sell your data.

2. Information we collect

Content you create (stored on your device). Recipes, formulas, notebook entries, grocery lists, photos you attach, and similar content you author are stored locally on your device using Apple's on-device storage. This content is not uploaded to our servers, whether or not you have an account, with one exception you trigger yourself: the organize request described next.

Drafts you ask us to organize (sent at your explicit action). The 清書 (fair copy) feature turns a rough draft into a typeset recipe page. When you tap the seal, the text of that draft is sent over an encrypted connection to our server (api.ramenlab.app) and passed to a large-language-model provider to be organized; the finished fair copy is returned to your device. Nothing is sent while you write: only tapping the seal sends the draft, and only that draft is sent. The draft is not stored on our servers and is excluded from our server logs and error reports (section 7). The model providers that process it are named in section 5, each with where it processes data. The request carries your authentication token, and our server keeps content-free, account-linked records of the request: its size and outcome, never its words (this is how daily limits are enforced and how we monitor the feature's health).

The name you give the app (stored on your device). During setup the app asks what to call you. That name is stored on your device, where it is used to greet you and to title the starter recipes, notebook entry, and reminders the app sets up for you. It is never transmitted to us or to anyone else, it is separate from any account you may create, and it is not attached to the crash data described below. You can change or clear it at any time in Settings → the Seal. Clearing it removes the stored name; titles that already contain it keep those titles until you rename them.

If you create an account (optional). You can sign in with Apple, Google, or Facebook. When you do, we receive and store: your email address, the name your sign-in provider shares with us (with Sign in with Apple you can hide both your real name and email), and a random account identifier. We use these solely to authenticate you, keep your access working across reinstalls, and associate your subscription. Authentication is operated through Supabase, our sign-in and database provider, and our own backend. We do not use your email for marketing, and creating an account does not upload the content you create — it stays on your device in this version.

A random on-device identifier. The app generates a random identifier (a UUID) the first time it runs and stores it on your device. It is used internally to tag the content you create as "yours" so your data stays consistent across the app. If you create an account, this association is what lets the app keep your content intact when you sign in; the content itself still does not leave the device.

Search queries. When you search the recipe library, your search text is sent to our server to return results. Searches of your own notebook and content run on your device only. We do not build advertising or marketing profiles from search queries.

Requests to our server. When the app talks to our backend (for example to search the library, check subscription status, or delete your account), the request carries standard technical information — such as your IP address and the app version — and, if you are signed in, your authentication token. We use this to operate the service, not to track you.

Crash and diagnostic data (via Sentry). When the app crashes or hits an error, we collect technical diagnostic information through Sentry (our crash and performance monitoring provider) to find and fix bugs. This typically includes the type of error, a stack trace, the app version, your iOS version, and your device model. We do not attach your name, email, or account to this data, and the app does not associate it with the identifiers described above. A small sample of performance traces (approximately 10%) is also collected to help us diagnose slowness.

Subscription status (via Apple). If you purchase Ramen Lab Pro, Apple processes the transaction and tells the app whether you have an active subscription so we can unlock paid features. See section 4.

Information you send to other websites you choose. If you use the "import from a link" feature and paste a recipe URL, the app fetches that web page directly from the website you chose — the same way a browser would — in order to read the recipe. That request goes to the third-party website, not to us, and is governed by that website's privacy practices. We do not log the URLs you import.

2a. The feedback board

Ramen Lab has a public feedback board at ramenlab.app/feedback where feature requests and known bugs can be read and upvoted. Three pieces of information are involved, and nothing else:

• What you write, when you send it. Requests are written inside the app (Settings → the suggestion box). Your words arrive privately. Nothing you write is published automatically — a person reads every submission, and what appears on the public board is a title and description we write. Your original text stays private.
• An optional email address. If you add one to a submission, it is stored so a person can write back to you by hand from support@ramenlab.app. It is not added to any mailing list, it triggers no automated email of any kind, and it is never shown on the public board.
• A random vote identifier. The first time you upvote something, a random identifier is generated and stored — in your browser's local storage on the web, or the on-device identifier described in section 2 in the app. It exists so one browser or one device counts as one vote. You can clear it at any time by clearing your browser storage.

How votes are counted, precisely. When you vote, our server combines that random identifier with your IP address and runs both through a one-way cryptographic function (a salted HMAC). Only that irreversible result is stored. We do not store your IP address with your vote, and we do not log it. The secret used in that function lives in our server configuration and never in the database, so the stored values cannot be turned back into an identifier or an address. This is also why vote counts are labelled "votes" and never "people": what we can honestly count is how many browsers or devices asked for something, which is not the same as how many humans did.

You do not need an account to read the board or to vote.

3. Information we do NOT collect

To be explicit, in this version of the app:

• No account is required. If you never sign in, we never receive your email, your name, or any account identifier. (Setup asks for a name so the app can address you; it stays on your device and is never sent to us — see section 2.)
• No advertising and no ad tracking. We do not show ads, we do not use the Advertising Identifier (IDFA), and the app does not present an App Tracking Transparency prompt because it does not track you.
• No product/usage analytics are sent from the app. The app includes an analytics library (PostHog), but it is turned off by default the moment the app launches, so no usage or behavioral analytics events are sent by the app in this version. When you use the organize feature, our server records content-free, account-linked analytics about the request (its size and outcome, never its words), as described in section 2.
• No selling or sharing of your data with data brokers or for cross-app advertising.
• No location, contacts, or health data collection.
• No passwords held by us. Sign-in happens through Apple, Google, or Facebook; we never see or store a password for you.

To be equally explicit about the one case content does leave the device: a draft you organize with the 清書 feature is sent to our server at your action, as section 2 describes. No other content you author leaves your device.

4. Payments and subscriptions

Ramen Lab offers an optional subscription, Ramen Lab Pro, sold through Apple's In-App Purchase system. All payments are processed by Apple. We never see or receive your credit card number or other payment details. The app only learns, from Apple, whether you currently have an active subscription, so it can unlock paid features. Your purchase history is handled by Apple under Apple's privacy policy.

5. Third parties and service providers

In this version, the app relies on the following third parties:

• Apple — provides on-device storage, the App Store, In-App Purchase (payment) processing, and Sign in with Apple. Apple's handling of your information is governed by Apple's privacy policy.
• Supabase — operates our sign-in (authentication) and account database. If you create an account, your email address, sign-in name, and account identifier are stored with Supabase.
• Vercel — hosts our backend server (api.ramenlab.app), which handles library search, subscription checks, and account deletion.
• Google / Facebook — only if you choose to sign in with them; they act as your sign-in provider under their own privacy policies, and they tell us your email and name (as you've configured with them).
• Sentry — receives the crash and diagnostic data described in section 2 to help us fix bugs and improve stability. Sentry processes this data in the United States.
• OpenRouter, only for drafts you organize — when you use the 清書 feature (section 2), the draft text is passed to OpenRouter (OpenRouter, Inc., United States), an AI model router, which forwards it to the large-language model that serves your request (models from DeepSeek, OpenAI, and Google). We send the draft text only, with no name, email, or account identifier attached, and every request instructs OpenRouter to route only to serving providers that do not retain and do not train on the text (OpenRouter's zero-data-retention routing). OpenRouter's own handling is governed by its privacy policy.
• Recipe websites you choose — only when you actively import a recipe from a link, as described in section 2.

We do not use third-party advertising networks or data brokers.

6. Data storage and iCloud

The content you create is stored locally on your device. A draft you organize with the 清書 feature is processed by our server and a model provider to produce your fair copy and is not stored on our servers; the fair copy it returns is stored on your device like everything else you author (section 7).

A future update will add optional iCloud sync so you can keep your content in sync across your own Apple devices. When that feature ships, sync will use your own private iCloud account (provided and controlled by Apple); your content would be stored in your iCloud, not on our servers, and you would control it through your Apple ID and device settings. iCloud sync is not active in this version — your content remains on the device you created it on, whether or not you sign in.

7. Data retention

• Content you create stays on your device until you delete it in the app or uninstall the app. Uninstalling the app removes its local data from your device.
• Account data (email, sign-in name, account identifier) is retained until you delete your account — which you can do in the app, immediately and permanently (section 8).
• Drafts you organize, and the fair copies made from them, are processed in memory to answer your request and are not stored on our servers; the draft text is excluded from our server logs and error reports. The content-free records of each request (size and outcome, tied to your account for daily limits) are retained on a rolling basis and pruned automatically. The model router named in section 5 is instructed on every request to use only serving providers that do not retain the text.
• Crash and diagnostic data is retained by Sentry for a limited period under Sentry's standard retention policy and then deleted.
• Subscription/purchase records are retained by Apple under Apple's policies.

8. Your rights and choices

• You control your content directly. The content you create is yours and lives on your device, and nothing you author is sent to us except a draft you explicitly organize with the 清書 seal (section 2). You can delete individual items in the app, and uninstalling the app removes its local data.
• Delete your account in the app. If you created an account, go to Settings → Account → Delete Account. Deletion is immediate and permanent: it removes your account and the data associated with it from our systems. Your on-device content is unaffected and remains on your device.
• Access, correction, or export requests. For requests relating to data we hold (for example, account data or crash diagnostics), contact us at support@ramenlab.app and we will respond consistent with applicable law.
• Subscriptions. You can view, manage, or cancel a subscription at any time in your Apple ID / App Store settings. Deleting your account does not by itself cancel an Apple subscription — manage that in your Apple settings.

9. Children's privacy

Ramen Lab is a general-audience cooking app and is not directed to children under 13 (or the equivalent minimum age in your jurisdiction). We do not knowingly collect personal information from children. If you believe a child has provided us information, contact support@ramenlab.app and we will address it.

10. International users

The app is available wherever Apple distributes it. The data described above is processed by our service providers (Supabase, Vercel, Sentry) in the United States and/or other regions where those providers operate. If you use the 清書 organize feature, the draft you send is additionally processed by OpenRouter (United States) and the model serving provider it routes your request to under the zero-data-retention restriction described in section 5. By using the app you understand that this processing may occur outside your country of residence.

11. Future changes to the app

We are actively developing Ramen Lab. Planned additions include optional iCloud sync and in-app data export. When new features involve new data handling, we will update this Privacy Policy and its effective date before those features become active.

12. Changes to this policy

We may update this Privacy Policy from time to time. When we make material changes, we will update the "Effective date" above and, where appropriate, notify you in the app or on our website. Your continued use of the app after an update means you accept the revised policy.

13. Contact

Questions or requests about this policy or your data:

• Privacy/legal: legal@ramenlab.app
• Support: support@ramenlab.app

Ramen Lab Apps — publisher of Ramen Lab.