Atlas Privacy Policy
App: Atlas (iOS) Provider: Nira, LLC, Nashville, Tennessee Contact: support@nira-health.com Effective date: June 12, 2026 Version: 2026-06-12-privacy
This policy explains what information the Atlas app collects, why we collect it, who else touches it, how long we keep it, and the choices you have. We wrote it to be read by parents, not lawyers. If anything is unclear, email us at support@nira-health.com and we will answer plainly.
This policy works together with our Terms of Service and our Children's Privacy (COPPA) Notice.
Who Atlas is for, and whose data this covers
Atlas is used by an adult parent or legal guardian. You create the account, and you add one or more children to track. Children do not have their own accounts, do not log in, and do not use the app. Everything Atlas knows about your child, you entered.
This policy covers two kinds of information:
- Your information — the account holder's own data.
- Your child's information — the information you provide about your child. Because this is information about children, additional protections apply. Our Children's Privacy Notice describes those protections and your rights as a parent in detail.
Where this policy applies (and where it doesn't)
This policy covers the Atlas iOS app. Information you enter in Atlas is consumer health information governed by this policy. It is not a clinical or medical record.
Some families also enroll in Nira Health's clinical program. If you do, the records created in your clinical care are governed by the clinic's HIPAA Notice of Privacy Practices, available at https://www.nira-health.com/privacy.html — not by this policy. Atlas data is not moved into your clinical record without your explicit authorization at enrollment.
The Nira Health website (www.nira-health.com) has its own privacy policies, including a consumer health data privacy policy that covers website visits and marketing. The website's practices are not the app's practices; this document describes only what the Atlas app does.
What we collect and why
| What | Details | Why we collect it |
|---|---|---|
| Your account info | Email address, first and last name, phone number (optional) | To create and secure your account, sign you in, and contact you about your account |
| Your child's profile | First name (last name optional), date of birth or age, sex (optional), an avatar you choose | To personalize tracking and make Audrey's guidance age-appropriate |
| Daily check-ins | Your 1–5 ratings of your child across five channels — Sleep, Mood, Cooperation, Focus, and Connection — plus your own "You" self-rating | The core of the product: tracking your child's wellbeing over time, and noticing when you are running on empty too |
| Logged events | A category, a severity level, a time, and an optional short note you write in your own words | To capture notable incidents and wins alongside the daily ratings |
| Computed scores and insights | A wellbeing score, per-channel trends, and patterns and insights derived from what you entered | The product's output: turning your entries into something you can see and act on |
| Audrey conversations | The messages you send to Audrey (our AI coaching assistant), Audrey's replies, a running summary of your past conversations, and internal safety flags | To generate coaching responses that actually know your family's week, and to keep the conversation safe |
| Subscription status | Whether you have an active subscription, via Apple In-App Purchase | To unlock the features you've paid for. Apple handles payment; we never see your card number |
| Consent records | Which version of the Terms, this policy, and the COPPA notice you accepted, and when | To prove we asked, and you agreed |
| Push notification token | A device token from Apple, only if you turn notifications on | To deliver reminders and notifications |
| Diagnostic data | Crash and error reports — device model, OS version, app version, what went wrong, and your account ID — sent to Sentry, our crash-reporting service | To find and fix bugs and crashes |
What we do not collect
- No photos, videos, or audio. Atlas does not access your camera, microphone, or photo library.
- No location data.
- No advertising identifiers. There are no advertising or analytics SDKs in the app, and the app does no cross-app tracking.
- Nothing from your child directly. All child data is entered by you.
Audrey and Anthropic — read this section
Audrey is an automated AI assistant. No live person monitors your conversations. Audrey is not a clinician, and Atlas cannot help in an emergency — if your child is in crisis, call 911, call or text 988, or contact your pediatrician.
Audrey's responses are generated by AI models from Anthropic (the maker of Claude). When you send Audrey a message, we send Anthropic your message together with context about your family, including:
- your child's first name, age, and sex;
- your child's current wellbeing scores and recent check-in ratings;
- recent logged events, including the notes you wrote;
- a running summary of your past conversations, and the recent turns of the current conversation.
We want to be direct about this: this information identifies your family and is not anonymized before it is sent. Audrey is useful precisely because she knows your child's week — and that requires sending this context to Anthropic to generate each response.
Anthropic processes this data as our service provider under a Business Associate Agreement and confidentiality terms with us, which require it to safeguard the data and keep it confidential. Under Anthropic's commercial API terms, data submitted through the API is not used to train Anthropic's models.
A few more things you should know:
- Crisis disclosures are handled specially. If a conversation touches on crisis, self-harm, abuse, or an acute medical situation, that content is excluded from the durable family summary Audrey keeps, and Audrey is instructed never to proactively bring a past crisis disclosure back up.
- Audrey only processes data when you use her. If you do not want your family's information sent to Anthropic, do not use the Audrey feature — the rest of Atlas (check-ins, events, insights) works without her, and none of it is sent to Anthropic unless you chat. Deleting your account deletes the conversation history and family summary we hold.
- We keep internal safety flags and usage records on Audrey conversations for quality and safety review, and we operate a kill switch that lets us disable Audrey across the service immediately if something is wrong.
Who else receives data, and what they receive
We do not sell your or your child's personal information. We do not share it with advertisers, data brokers, insurance companies, or employers. We use a small number of service providers to run Atlas. Here is the complete list and exactly what each one receives:
| Service provider | What they do for us | What they receive |
|---|---|---|
| Supabase | Our database, authentication, and file storage, hosted on AWS in the United States (us-east-1) | Everything Atlas stores (see the table above). Supabase operates under a Business Associate Agreement with us, and your family's data is isolated by database-level security rules |
| Anthropic | Generates Audrey's responses | Your Audrey messages plus the family context described in the Audrey section above — child first name, age, sex, scores, recent events and notes, and conversation summaries. Handled under a Business Associate Agreement and confidentiality terms with us; not used to train Anthropic's models |
| Apple | App distribution (App Store / TestFlight), subscription billing (In-App Purchase), and push notification delivery (APNs) | Purchase and subscription information, and your device's push token. Apple's own privacy policy governs your Apple account; we never receive your payment card details |
| Sentry | Crash and error reporting (sentry.io) | Crash and error reports including device model, OS version, app version, and technical details of the failure, linked to your account ID so we can investigate problems you report. We configure Sentry not to collect personal details automatically, and we do not intentionally send any health data or child data to it |
| Resend | Sends transactional account emails, such as password resets | Your email address and the content of the account email being delivered. No child data, no health data |
| Cal.com | Optional scheduling, only if you choose to book a discovery call with Nira Health | If — and only if — you tap a booking link, you leave the app and enter your name, email, and a time on Cal.com's site, governed by Cal.com's privacy policy. We do not pass your child's health data into the booking |
That is the whole list. No data leaves these flows.
How long we keep data, and how deletion works
While your account is active, we keep your data so the product works: tracking only makes sense if history is retained.
When you delete your account (Profile → Delete Account, inside the app):
- Your household's data — your account info, your children's profiles, all check-ins, events, scores, insights, Audrey conversations, and the family summary — is deleted from our live systems immediately.
- Uploaded files are deleted immediately and are unrecoverable.
- Residual copies in our encrypted daily backups are purged within 7 days.
- We retain a minimal administrative record that a deletion occurred (it does not contain your or your child's data) so we can prove the deletion happened, along with the consent records the law requires us to keep.
Service providers' copies: deletion from our systems does not instantly erase data our service providers already processed. Anthropic and Sentry retain data for limited periods under their own terms before it is deleted on their side; Apple retains your purchase records under its own policies. We do not retain any path back to your deleted data.
Two practical notes: deleting your account does not cancel an active Apple subscription — manage that in your Apple account settings — and deletion is permanent. We cannot restore a deleted household.
Your rights and choices
- Review. Everything Atlas knows is visible in the app — your entries, your child's tracking history, scores, insights, and your Audrey conversations.
- Delete. In-app account deletion (described above) removes your household and all child data.
- Refuse further collection. Atlas only collects what you enter. You can stop further collection at any time by discontinuing use, and you can stop AI processing by not using Audrey.
- Notifications. Push notifications are off unless you opt in, and you can turn them off in iOS Settings at any time.
- Anything else. Email support@nira-health.com to exercise any right described in this policy or in the Children's Privacy Notice — including reviewing, correcting, or deleting information about your child. We will verify that you are the account holder before acting.
Parents' rights regarding children's information — including the right to review what we have collected about your child, refuse further collection, and direct deletion — are described in full in our Children's Privacy Notice.
State privacy rights
Depending on where you live, state law may give you additional rights over consumer health data — including rights to access, correct, and delete it, to receive a list of third parties that have received it, and to withdraw consent for its collection or sharing. The information in this policy (the collection table and the service-provider table above) is the complete picture of what we collect and share. To exercise any state-law right, email support@nira-health.com; if we decline a request, we will explain why and how to appeal.
How we protect your data
- Encryption. Data is encrypted in transit (TLS) and at rest.
- Family isolation. Database row-level security restricts every record to the household that owns it — your data is structurally inaccessible to other families.
- Access controls. Access to production systems is limited and credentialed; AI API keys live only on our servers, never in the app.
- Monitoring and a kill switch. We monitor for errors and abuse, and we can shut off the AI assistant service-wide immediately if a safety or security problem arises.
No system is perfectly secure, and we won't pretend otherwise. If a breach of your unsecured personal information occurs, we will notify you and the appropriate regulators as required by applicable law, without unreasonable delay, and we will tell you plainly what happened, what data was involved, and what we are doing about it.
Children's privacy
Atlas collects personal information about children under 13 (and minors generally), entered by their parent or guardian. We treat this as the most sensitive data we hold. We do not collect information from children directly, we do not use children's data for advertising or marketing of any kind, and we do not sell it — ever. The complete description of our children's-data practices, the parental consent we require, and your rights as a parent is in our Children's Privacy Notice, which is part of our agreement with you.
Changes to this policy
If we change this policy, we will update the effective date and version above. For material changes, the app will show you the updated policy and ask you to accept it again before you continue using Atlas — your previously recorded consent identifies the exact version you agreed to. For material changes to how we collect, use, or share children's information, we will obtain your fresh consent; continued use alone is not treated as agreement to those changes.
Contact us
Nira, LLC — Nashville, Tennessee support@nira-health.com
Questions, concerns, or requests about your or your child's data all go to the same place, and a person will read them.