How Peanut handles data
Peanut exists to solve a narrow, practical problem: figuring out what to cook with the food you already have.
To do that well, Peanut needs to pay attention to what happens inside the app.
Nothing more.
What Peanut pays attention to
Peanut keeps track of things that directly affect meal planning:
- meals that are suggested
- meals that are cooked
- meals that are skipped
- ingredients in your pantry
- changes you make to meals or recipes
- ratings or feedback you choose to give
This is how Peanut learns what works for your household.
When something gets cooked, Peanut remembers.
When something gets skipped, Peanut notices.
When you change a meal, Peanut learns from that.
All learning happens at the household level. Peanut is interested in patterns, not people.
What Peanut does not collect
Peanut does not collect or process special category or sensitive personal data.
This includes, but is not limited to:
- health conditions or medical diagnoses
- biometric data
- genetic data
- exact ages of children
- real names (placeholders work just fine)
- precise location data
- financial account or payment information
- grocery receipts
- photos of food (uploads are optional and user-initiated)
Preferences such as “avoid mushrooms” or “keep costs low” are treated as cooking preferences, not medical, financial, or personal classifications.
How identity works
Peanut is a single-user app.
You can create profiles for family members, but they do not need real names. Age is stored only as a broad category (child, teen, adult).
Peanut does not try to identify people. It only needs to understand who meals are for.
How learning works
Peanut learns from your household only.
Your meal history and preferences are not pooled with other households and are not used to build advertising profiles or shared recommendation models.
What Peanut learns about your kitchen stays with your kitchen.
Where your data lives
Peanut runs entirely on infrastructure we operate and control.
The backend, database, authentication system, and learning services are all hosted on our own servers.
Your household data does not pass through advertising systems, analytics platforms, or third-party tracking tools.
The only time data leaves our servers is when a limited subset is sent to the OpenAI API to generate suggestions or interpret changes.
This data is used only to return a response and is not used to train external models.
Security
We protect your data using industry-standard technical and organizational measures.
This includes access controls, encrypted storage where appropriate, and strict limitations on who can access production systems.
Free text and notes
Some parts of Peanut allow you to write text (for example, when modifying a recipe).
Whatever you choose to write is stored as part of that interaction. Peanut does not attempt to interpret or classify free text beyond its immediate use.
Your control
You can review, edit, or correct your pantry items, meals, and preferences at any time within the app.
You can delete meals, reset preferences, or remove your account entirely.
If you leave
If you delete your account, your household data is removed from active use.
Encrypted backups may persist for up to 90 days before being automatically overwritten or deleted. These backups are not accessed once an account is closed.
Human access
Peanut is designed to run automatically.
In rare cases, a human may access account data to investigate bugs, handle support requests, or fix broken behaviour.
There is no routine browsing of household data. Access is limited, purposeful, and infrequent.
Questions or concerns
If you have questions about privacy, data handling, or deletion, you can contact us at:
privacy@milkstack.com
The line we don’t cross
Peanut will never sell household data.
It will never profile children.
It will never infer health conditions.
It will never track you outside the app.
Peanut pays attention to dinner.
Not to you.
Comments
Displaying 0 of 0 comments ( View all | Add Comment )
Comments are moderated before appearing.