Version 1.0 · Effective 11 November 2026
1. Who we are
Everyday Flare is an app that gives parents and carers practical, OT-informed ideas for everyday life with their child. It is operated by Christopher Owen Armstrong, a sole trader trading as Flare Therapy (ABN 11 741 313 662), based in Western Australia. In this policy "we" and "us" mean Flare Therapy. Chris runs Everyday Flare and handles privacy questions himself.
Contact: Chris@Flaretherapy.com
We have written this policy around the Australian Privacy Principles.
2. Who can use Everyday Flare in the beta
The beta is for invited parents, guardians and authorised carers aged 18 or over who live in Australia. Before you start, you confirm: "I'm this child's parent, guardian or authorised carer, and I'm 18 or over." That is a statement you make; we don't check it. During the beta each household has one adult account. Children do not have their own accounts or their own spaces, and the app can't be used without signing in.
If you'd rather not agree: you don't have to. Before you start, we ask you to tick two boxes: that you're the child's parent, guardian or authorised carer and 18 or over, and that you agree to us collecting and using your child's information as this policy explains. If you don't tick them, choose Sign out: nothing new is saved, and nothing in your account changes. You can still ask for help with your account, or ask for it to be closed, without agreeing to anything (choose "Help with my account, or close it", or email Chris@Flaretherapy.com). You can't use the app itself without agreeing.
The beta is planned to run for about four weeks. Before it ends we'll tell you when it is ending and what happens to your access. Your information isn't deleted just because the beta has ended (see sections 8 and 9).
3. What we collect
Your account
- Your email address, used to sign you in with a one-time code sent to it.
- Your name as you enter it, your household and who belongs to it.
- That you accepted the "Before you start" notice (including your consent), and which version. If you close your account, that acceptance is marked as withdrawn (it isn't deleted), and you're asked to accept again if you reopen.
About your child — what you choose to add
- Name, date of birth or age, interests, favourite characters, communication, sensory preferences, triggers, what helps, things to avoid, what you're working towards, and the people who support your child (for example a teacher or therapist) if you add them.
- Some of this may be about your child's health, development, behaviour or wellbeing. We treat all of it as sensitive. You choose what to add, and you can leave any of it out. Before you start, we ask you to agree to us collecting and using it, and to sending the relevant parts to OpenAI in the United States (section 5).
- Photo uploads (for supports and a profile picture) are switched off during the beta, so we don't collect photos.
What you do in the app
- The situations you describe and the questions you answer, the supports Flare makes and the ones you save, and which ideas you chose to try and how you said they went.
- Build Skills: the skills you choose to work on, how much help your child needs with each part, the plan you're following, and the practice you record. These come from fixed choices; Build Skills has no free-text boxes.
- When you tell Flare an idea didn't help or doesn't suit your child, your words (up to 240 characters) are saved as a note that Flare takes into account when suggesting ideas for that kind of situation. If you clearly say a kind of idea doesn't suit your child at all (for example "he hates sticker charts"), Flare leaves that kind of idea out for every situation and tells you it has done so. You can narrow it to one situation, or stop it, in What Flare has learned. If it isn't clear which you mean, Flare asks you.
- When an idea didn't help with something, Flare leaves it out for that kind of situation until you say otherwise. After about three months it may ask once whether you'd like it to include that idea again; your answer is saved. If a similar situation comes up, Flare asks you before leaving the idea out there too.
- If you tell Flare in your own words how an idea went, those words (up to 240 characters) are saved with that report.
- What you ask Flare about: when you ask Flare for ideas, a short note is saved automatically, once per conversation and topic: your own first words for that request (up to 240 characters) and the topic as Flare read it. It appears in Progress & insights under "What you've asked Flare about", where you can change the topic or choose "Stop using this". If Flare's safety rules notice anything in your words, the words are not kept, only the topic. Once a safety message has been shown in a conversation, no new note is saved from that conversation. A note saved earlier in the same conversation, before the safety message, is not removed: it stays (with up to 240 characters of your words) until you choose "Stop using this". A note does not mean the situation happened more than once, and notes are not counted as repeated incidents in Progress. They are separate from safety records (below), and neither is a transcript.
- Everyday Flare does not save a transcript of your conversations.
Flare's own notes (its interpretations)
- Its reading of each situation you describe — for example, what may have made it harder and how intense it was.
- A short summary, written by the AI, of what you asked Flare for (up to 120 characters), saved with the ideas Flare offered you.
- The topic of each Ask Flare request, as Flare read it.
- These are Flare's interpretations, not facts about your child. They are used to suggest ideas, and the situation readings are also used to show patterns in Progress & insights, where they are labelled as Flare's reading.
Safety concerns
- If something you write raises a safety concern, we keep a protected safety record that a safety message was shown: which kind of concern, when, and where in the app.
- Most safety records keep only that. This "category-only" record holds the kind of concern and a few fixed labels, never your words or the message shown. It is used whenever Flare's built-in safety rules have not confirmed the concern, and whenever Flare cannot tell who the concern is about (the record then notes "person involved: unclear").
- A record keeps your exact words only when the built-in rules confirm a serious concern of one of these kinds about your child or you: suicide risk, self-harm, abuse or neglect, family violence, or your own crisis. It then also keeps the safety message you were shown. Before anything you type is saved or sent anywhere, card numbers, passwords and one-time codes are replaced with "[payment or security detail removed]". Where the screen showed "call 000", the stored message says "call emergency services".
- When only the AI check finds a concern, the record keeps only the category "general concern": never your words, the AI's reading, or the AI-written safety message.
- Who can see them. Safety records are not shown in the app, are kept separately from what Flare learns about your child, and are not used to make suggestions or sent to the AI with later requests. Chris is the only person with administrator access, and would look at them only if a safety, legal or complaint need arises. Inside the app, during the beta, only you can reach your child's records.
- If saving one fails (for example, your connection drops), the safety message is still shown straight away, and the app tries to save the record again while it stays open. It is never stored in your browser.
- How long they are kept has not been decided yet (see section 8). Closing your account does not delete them.
Feedback and problem reports
- If you send feedback or report a problem in the app, we save your message with your account and household. A problem report also notes the screen you were on and the app version. Your child's details and your conversations are not attached.
- If you email us, we receive whatever you include. An email doesn't prove who you are, so we confirm a request with the email address your account uses before acting on it (section 10).
- End-of-beta questionnaire: we may ask you to fill in an optional questionnaire when the beta ends. If we send one, we'll say which tool it uses and what it collects when we send it.
If you use the microphone
- Speaking instead of typing uses your phone's or browser's own speech recognition, which turns your speech into text. Depending on your device and browser, that service may be run by the browser maker (for example Google for Chrome, or Apple for Safari) under their own terms. Everyday Flare receives only the text, never stores audio, and sends nothing until you press send.
Product and technical information
- Simple usage events (for example, that a session started or a feature was opened), made of fixed labels — never your words.
- Counters that limit how often the AI features can be used.
- Your browser keeps your sign-in session in its own storage so you stay signed in. The app saves nothing else in your browser. We do not use advertising cookies or third-party analytics.
- Our hosting providers keep standard technical logs (such as IP address and time of request) under their own terms. On our current plans, Supabase keeps its request logs for 1 day. Resend keeps a record of each email it sends for 30 days: your sign-in code emails (including your email address) and the alerts to Chris.
4. Why we use it
- To provide the app: sign-in, your child's profile, ideas and supports, Build Skills plans, and remembering what helped.
- To respond to safety concerns with the right safety information.
- To answer feedback and problem reports, and to fix and improve the beta.
- To keep the service secure and prevent misuse (for example, usage limits).
We do not sell personal information or use it for advertising.
5. How Flare uses AI
What the AI does. Everyday Flare uses an AI service, OpenAI, to:
- read what you describe (for example, what the situation is about and what may have made it harder);
- write Flare's replies, its ideas, step-by-step help and supports, and the pictures in supports;
- help check messages for safety concerns, and check Flare's own wording before you see it.
Where Flare's ideas come from. Some ideas come from a library of strategies that Chris, an occupational therapist, has reviewed and approved word for word. Others are written by the AI for your situation, within rules Flare sets. For example:
- during a hard moment, only calming ideas;
- nothing you've said didn't help or doesn't suit your child;
- nothing outside everyday support.
A reviewed strategy is only used when it closely fits what you described. AI-written ideas are not individually checked by a person before you see them. All ideas are general suggestions, not therapy or professional advice for your child.
What is sent to OpenAI. Through Everyday Flare's own server:
- your messages;
- your child's first name and age;
- for some features, details from your child's profile (interests, favourite characters, communication, sensory preferences, triggers, what helps and things to avoid). Help Right Now's replies send these details each time. Ask Flare sends only the details that fit what you've asked about:
- things to avoid, always;
- how your child communicates (a typed description only when it is about communication);
- triggers that match your request;
- up to three things that help, leaving out any kind of idea you've said doesn't suit;
- sensory preferences, for tasks like dressing, teeth or bath time;
- interests and favourite characters, for getting started or routine tasks, never during a meltdown. Under the ideas, Flare tells you which parts of the profile it used. The people who support your child are never sent.
- for some features, short notes of how earlier ideas went, including your own words about what didn't help or doesn't suit your child;
- in Ask Flare, earlier messages from the same conversation, so Flare can follow it.
Safety check and ordinary reading. Your messages are sent both to check them for safety concerns and to write replies. In Ask Flare, to keep replies quick, Flare starts its ordinary reading of your message (with up to three of your earlier messages in that conversation) while the safety check is still running. It doesn't do this if Flare's built-in safety rules have already noticed something in your words. If the safety check then finds a concern:
- that reading is not used;
- nothing from it is saved;
- you see the safety message instead.
Your message has still been sent to OpenAI for that reading.
Your child's date of birth is never sent. Where you can, leave out other people's full names and identifying details.
What OpenAI keeps. OpenAI keeps and uses what it receives under its own terms. We can't see OpenAI's logs, so we can't confirm what it keeps or for how long. What we can say:
- The data-sharing settings on our OpenAI account are switched off. Under OpenAI's terms, that means it doesn't use these messages to train its models.
- Everyday Flare's requests for text replies ask OpenAI not to store them for later retrieval.
- OpenAI's terms say it may still keep the messages it receives for up to 30 days to monitor for abuse, or longer if the law requires. We can't switch this off.
- OpenAI processes these messages outside Australia, in the United States.
Safety checks have limits. Flare checks what you type when you describe a situation, in Help Right Now and Ask Flare, for safety concerns, using fixed safety rules first and then the AI. When Flare's built-in safety checks don't settle what to show, an AI service reads your message and may write the safety message you see. Flare adds the 000 button and verified support numbers. No person checks the AI-written wording before it appears, and it can be wrong.
If you reply to a safety message, the AI service may receive your reply along with up to three earlier exchanges, each containing a parent message and Flare's reply. The safety check itself doesn't attach your child's saved name, age or profile. If you type those details into a message, they may still be sent as part of that message.
Words you type into a support (for example First / Then, choices, or a change you ask for) are checked on your device against the fixed rules only. It can miss things. Feedback, problem reports and the details in your child's profile are not checked. If anyone is in danger, call 000.
6. Who we share it with
We use these service providers to run Everyday Flare. They handle information for us:
| Provider | What for | Where |
|---|---|---|
| Supabase | Database, sign-in, file storage and server functions | Singapore (our database region) |
| OpenAI | Reading messages, writing replies, ideas and supports, checking safety and wording, generating support pictures | United States |
| Resend | Sending your sign-in codes to your email address, and sending Chris an alert when feedback arrives (the alert contains only the type, area and time, never your message) | Japan |
| Cloudflare | Hosting the app | Cloudflare's worldwide network: usually a server near you, which may be outside Australia |
| Google Fonts, jsDelivr, unpkg | Delivering fonts and app code to your browser (they see your IP address) | Worldwide networks: usually a server near you, which may be outside Australia |
| GitHub (Microsoft) | Running the weekly backup job. Your information is there only in the job's memory for a few minutes, before it is encrypted | United States |
| Cloudflare R2 | Storing encrypted backups, which Cloudflare can't read | Oceania |
| healthchecks.io | Alerting Chris if a backup fails or is missed. It receives counts and codes only, never family information | European Union |
Some of these providers process information outside Australia, as shown in the table.
We don't share your information with anyone else unless the law requires or allows it.
Sharing you choose to do: you can print or share supports, skill steps and other resources Flare makes — for example with your child's teacher, school or therapist. Once you share something, how the person you shared it with handles it is up to them, so share only what you're comfortable with.
7. Who at Everyday Flare can see it
Inside the app: only people in your household can see your household's information, and only for children they have access to. During the beta that is you, because each household has one adult account.
At Everyday Flare: Chris, who runs Everyday Flare, is the only person with administrator access to beta data. There is no other staff, and no clinical adviser or anyone else sees it. Chris reviews feedback and problem reports in Everyday Flare's database dashboard, which needs his own sign-in. He looks at a family's other information only to answer a request from that family, to fix a problem they report, to deal with a safety, security or legal need, or where the law requires it.
Our service providers (section 6) host and process information for us. They can access it only as needed to provide their service, under their own terms.
8. How long we keep it
- Your account and child information: kept while your account is open, and while it is closed (see section 9). Nothing is deleted automatically during the beta.
- Feedback and problem reports: kept with your account. How long they are kept after the beta has not been set yet, and nothing is deleted automatically during the beta. We'll update this policy before any of them are deleted.
- Safety records: kept with your child's information; how long has not been decided. Nothing is deleted automatically during the beta.
- Feedback alert emails stay in Chris's mailbox.
- Backups: each week an automated job copies the database (including sign-in details) and stored files, encrypts them so only Everyday Flare's owner can open them, and stores them with Cloudflare (R2) in Oceania for up to 5 weeks. The job runs on GitHub's servers, normally in the United States, where your information exists unencrypted only in the job's memory for a few minutes. Something deleted from the app (for example a saved support) can remain in a backup until that backup is deleted. If we ever had to restore the app from a backup, it would go back to how it was when that backup was made, and anything you deleted or changed since could come back. We would keep the app closed to families until every account closed since that backup is closed again. If we couldn't tell which accounts were closed, we would close every account, and you could reopen yours by signing in and choosing Keep my account.
- Providers: information sent to OpenAI, Resend, Cloudflare and others is kept under their own terms (sections 5–6).
9. Closing your account
You can close your account at any time from More → Privacy & account → Close my account. It closes straight away: while it's closed, no one in your household can see your children's information, and nothing is deleted. You can restore it whenever you like by signing in and choosing Keep my account.
You can also ask Chris to close it for you by emailing Chris@Flaretherapy.com (you don't need to accept the "Before you start" notice to ask). Chris handles these requests personally.
Permanent deletion is not switched on during the beta. A closed account stays closed rather than being erased. If you ask for your information to be permanently deleted (email Chris@Flaretherapy.com), Chris will explain what can and can't be done, record your request, tell you when he'll next write to you with an update (a date for an update, not for deletion), and ask whether you'd like your account closed. He closes it only if you say yes. Closing stops your household's access and stops your information being used, but doesn't erase it. During the beta, the only information that can be permanently deleted is saved supports, which you can delete yourself in the app. Everything else stays stored, and we can't promise permanent deletion or a time frame for it. If permanent deletion becomes available, Chris will contact you before anything is deleted, to check you still want it, and this policy and the "Before you start" notice will be updated first.
If you reopen a closed account, you'll be asked to agree to the "Before you start" notice again before you can use the app, and nothing you write is sent to the AI service until you do.
10. Your choices and rights
- See and correct: you can view and edit your child's profile, and view what Flare has learned about your child (and print it).
- Stop Flare using something: in What Flare has learned, each report of how an idea went and each note about what doesn't suit your child has "Stop using this". Flare then stops using it when suggesting ideas and stops showing it.
- It is not permanently deleted — it stays in your account's records.
- If you reported on the same idea more than once, Flare goes by your newest remaining report, so stopping the newest may mean an older one is used instead.
- Flare's own readings of situations don't have "Stop using this".
- The notes of what you've asked Flare about (Progress & insights) have "Stop using this" and "Change topic".
- Your child's profile details are yours to edit or remove at any time, and Flare only uses what is there.
- Delete: you can delete saved supports in the app. Other information can't be permanently deleted in the app yet, for example Flare's own notes, reports and notes you've stopped Flare using, Build Skills practice records, and the Sensory Snapshot.
- Withdrawing consent: you can withdraw your consent and stop using Everyday Flare at any time. To stop your information being used, close your account yourself (section 9), or email Chris@Flaretherapy.com and he will close it for you and confirm by email. While it's closed, your information isn't used in the app or sent to the AI service, and nobody in your household can see it. Nothing is deleted: it stays stored, including in our weekly backups, and you can restore it by signing in and choosing Keep my account.
- Ask us: email Chris@Flaretherapy.com to ask for a copy of your information, to correct it, to close your account, to ask about deletion (section 9), or with any privacy question. Chris handles every privacy request personally and aims to reply within five working days. Email isn't monitored for urgent help.
- Checking it's you: before acting on a request, Chris checks it comes from the email address your account uses. He then writes to that address (a new email, not a reply) and acts once you confirm from it. If you no longer have that email address, Chris can't confirm the request is yours, so he won't act on it. Nobody can sign in without that address either.
- Complaints: if you're unhappy with how we've handled your information, email us first and we'll try to put it right. If you're still not satisfied, you may be able to complain to the Office of the Australian Information Commissioner (OAIC, oaic.gov.au).
11. Security
We use access rules in the database, so each household can reach only its own information; private file storage; sign-in without passwords; and separate keys for our server functions. No system is perfectly secure.
If we find that your information may have been seen, changed, lost or shared by someone who shouldn't have it, Chris will look into it straight away and act to stop it. If it's likely to cause you or your child serious harm, he will tell you what happened, what information was involved, what has been done, and what you can do. Where the law requires it, he will also tell the Office of the Australian Information Commissioner.
12. Changes to this policy
If we change how we handle your information in a way that matters, we'll update this policy and ask you to read the updated "Before you start" notice in the app.