PRIVACY POLICY
This Privacy Notice for Reformic OU ("Reformic," "we," "us," or "our") explains how personal information is processed when you use Echologue, our mobile application, or visit a website that links to this notice (together, the "Services").
Echologue is designed as a local-first application. Your permanent journal archive is stored on your device. Certain optional features require temporary transmission of information to third-party service providers, including artificial intelligence providers. Echologue asks for your permission before personal journal, voice, or chat content is sent for third-party AI processing.
Questions or concerns? Contact us at reformic.est@gmail.com.
SUMMARY OF KEY POINTS
Where is my journal stored? Your entries, structured journal data, local semantic indexes, remembered chats, memory profile, and recent-state information are stored in Echologue's local database on your device. We do not operate a permanent server-side journal archive.
What leaves my device? Only information needed for a feature you request may be transmitted, such as a voice recording for transcription, journal text for organization or embeddings, or a chat question together with selected relevant journal context for an answer.
Who receives it? Requests pass through Echologue's Cloudflare-hosted API and may be processed by Groq for transcription, and by OpenRouter and its selected AI inference providers for entry organization, embeddings, and generated answers. Subscription information is processed by Apple or Google and RevenueCat.
Do I have a choice? Yes. Before third-party AI processing begins, Echologue presents a disclosure and asks you to affirmatively allow it. You may decline or later withdraw permission in Settings. AI-dependent features will then remain disabled, while locally available features and your on-device archive remain available.
Is my content used to train AI models? We do not authorize service providers to use your journal, voice, or chat content to train public or general-purpose AI models. We use privacy controls, including zero-data-retention routing for OpenRouter requests, where configured and available.
Could sensitive information be involved? Yes. A private journal may contain health, relationship, sexual, religious, political, financial, or other sensitive information that you choose to enter. We do not ask you to provide any particular category, but content you submit may be sensitive.
What controls do I have? You can review, edit, export, or wipe your local archive in the app and can revoke AI-processing permission at any time.
TABLE OF CONTENTS
- WHAT INFORMATION IS PROCESSED?
- WHERE YOUR JOURNAL IS STORED
- THIRD-PARTY AI PROCESSING AND CONSENT
- WHAT DATA IS SENT, TO WHOM, AND WHY
- SUBSCRIPTIONS AND PURCHASES
- WEBSITE AND TECHNICAL INFORMATION
- LEGAL BASES FOR PROCESSING
- HOW LONG INFORMATION IS KEPT
- HOW WE PROTECT INFORMATION
- INTERNATIONAL DATA TRANSFERS
- CHILDREN'S PRIVACY
- YOUR PRIVACY RIGHTS AND CONTROLS
- UNITED STATES PRIVACY DISCLOSURES
- OTHER REGIONAL RIGHTS
- DO-NOT-TRACK SIGNALS
- CHANGES TO THIS NOTICE
- HOW TO CONTACT US
- HOW TO REVIEW, EXPORT, OR DELETE YOUR DATA
1. WHAT INFORMATION IS PROCESSED?
In Short: Echologue processes information you create in the app, device and purchase information needed to operate the Services, and limited technical information.
Journal and chat content you provide. This may include typed entries, voice recordings, transcripts, dates, optional location information, names, places, events, feelings, tags, themes, questions, corrections, preferences, remembered chat messages, and any other information you choose to record.
Derived journal information. Echologue may create structured facts, feelings, entities, people, tags, semantic vectors, retrieval results, memory summaries, and recent-state summaries from your entries and remembered chats. These are used to organize and retrieve your own history.
Audio and microphone access. Echologue requests microphone permission so you can record a voice entry. Recording begins only when you choose to use the recording feature. Audio is stored locally while it is being handled and may be sent for transcription if you have allowed third-party AI processing.
Purchase and entitlement information. When you purchase or restore a subscription, Apple App Store or Google Play and RevenueCat may process purchase receipts, product identifiers, transaction status, subscription status, an app-specific user identifier, device or platform information, and related technical information. We do not receive or store your full payment-card number.
Technical information. Our infrastructure and service providers may process limited operational information such as timestamps, IP address, request identifiers, device or app version, model requested, token or audio duration counts, latency, error information, and security or abuse-prevention signals. We do not use journal content for advertising or cross-app tracking.
2. WHERE YOUR JOURNAL IS STORED
In Short: The permanent Echologue archive is local to your device.
Echologue stores your journal entries, structured fields, semantic vectors, remembered chats, local memory profile, recent-state summary, app settings, import records, usage counters, and entitlement state in a local database on your device. Retrieval and selection of relevant entries occur locally.
We do not maintain a cloud account containing a copy of your permanent journal archive and do not provide automatic server-side journal backup. Your device's operating-system backup features may separately back up app data depending on your device and account settings; those backups are controlled by Apple, Google, or your chosen backup provider.
3. THIRD-PARTY AI PROCESSING AND CONSENT
In Short: Echologue asks for explicit permission before transmitting personal content to third-party AI services.
Before Echologue sends voice, journal, or chat content to a third-party AI service, the app displays a disclosure explaining:
- what categories of information may be sent;
- which service providers may receive it;
- the purposes of the transmission; and
- that permission can be declined or withdrawn.
You must take an affirmative action, such as tapping "Allow AI Processing," before this transmission begins. Merely installing or opening the app does not constitute permission.
If you select "Not Now" or withdraw permission in Settings, Echologue blocks new third-party AI requests, including transcription, remote entry organization, remote embedding generation, and AI-generated chat answers. Entries already stored locally remain on your device. Pending AI processing remains paused unless and until permission is granted again.
Microphone permission granted through your operating system is separate from permission to share information with third-party AI services. Granting microphone access does not by itself authorize AI transmission.
4. WHAT DATA IS SENT, TO WHOM, AND WHY
In Short: We disclose the specific data flows used to provide remote features.
| Information sent | Recipient | Purpose |
|---|---|---|
| Voice recording and limited request metadata | Echologue's Cloudflare-hosted API; Groq | Convert the recording into a text transcript. |
| Journal text and related entry context, which may include dates, names, places, facts, feelings, tags, language, and optional location information | Echologue's Cloudflare-hosted API; OpenRouter; the selected AI inference provider | Organize an entry into structured fields and themes. |
| Text derived from an entry or remembered chat | Echologue's Cloudflare-hosted API; OpenRouter; the selected embedding inference provider | Create numerical vector embeddings that are returned to the app and stored locally for semantic retrieval. |
| Your chat question, selected relevant journal excerpts, and relevant profile or recent-state context | Echologue's Cloudflare-hosted API; OpenRouter; the selected AI inference provider | Generate an answer grounded in your locally selected history. |
| Request metadata such as timestamps, request identifiers, model, usage counts, latency, errors, IP address, and security signals | Cloudflare, Groq, OpenRouter, and applicable inference providers | Deliver requests, authenticate access, measure usage, prevent abuse, diagnose errors, and protect service reliability. |
Cloudflare. Echologue uses a Cloudflare-hosted worker as an authenticated relay between the app and remote service providers. The worker receives request content in transit so it can forward the request and return the response. It is not used as a permanent journal database. Cloudflare may process limited network and security information under its own terms and privacy commitments. See Cloudflare's Privacy Policy.
Groq. Groq processes voice recordings for speech transcription. We do not enable optional persistence features for journal audio and do not authorize Groq to use inputs or outputs to train models. Groq may process limited usage metadata and applies its then-current data controls and retention terms. See Groq's data-handling documentation and Privacy Policy.
OpenRouter and AI inference providers. OpenRouter routes requests for entry organization, vector embeddings, and generated answers. We configure OpenRouter requests to deny data collection and restrict routing to zero-data-retention endpoints where supported. OpenRouter may route a request to an endpoint operated by a provider associated with the selected model. Current model families used by Echologue may be provided by or associated with Google, OpenAI, and Anthropic; the actual endpoint operator can vary based on the configured routing and availability. See OpenRouter's Privacy Policy, Zero Data Retention documentation, and provider-retention documentation.
We do not sell journal, voice, or chat content. We do not permit it to be used for targeted advertising. We do not authorize providers to use it to train public or general-purpose AI models.
Equivalent protection. We select and configure service providers so that personal content is subject to purpose limitations, security measures, retention controls, and restrictions on advertising and model training that are the same as or equivalent to the protections described in this notice. Providers may process content only to deliver the requested service, protect service reliability and security, comply with law, or as otherwise disclosed here. If we cannot obtain or configure appropriate protection for journal content, we will not intentionally route that content to the provider.
For information about the current model-provider families used through OpenRouter, see Google's Privacy Policy, OpenAI's Privacy Policy, and Anthropic's Privacy Policy.
Even when content retention is disabled, providers may retain limited non-content operational metadata, such as usage counts, latency, request identifiers, billing records, fraud-prevention signals, or security logs. Providers may also preserve information when legally required. Their current policies and technical controls govern those activities.
5. SUBSCRIPTIONS AND PURCHASES
In Short: Purchases are handled by the app stores and RevenueCat.
Echologue uses RevenueCat to manage subscription entitlements and restore purchases. Apple or Google processes the payment. RevenueCat may receive app-store transaction information, product identifiers, subscription status, an app-specific or anonymous user identifier, device and platform information, and related diagnostic information needed to provide entitlement services.
We use this information to determine whether premium features should be available, prevent purchase fraud, provide customer support, and maintain subscription functionality. For more information, see RevenueCat's Privacy Policy, Apple's Privacy Policy, and Google's Privacy Policy.
6. WEBSITE AND TECHNICAL INFORMATION
In Short: Our website may use basic analytics and hosting logs.
When you visit an Echologue or Reformic website, hosting, content-delivery, and analytics providers may process information such as your IP address, browser type, device type, approximate region, referring page, pages viewed, button clicks, and timestamps. This information is used to operate the website, understand aggregate traffic, diagnose failures, and measure actions such as app-store link clicks.
The website may use Google Analytics or Firebase Analytics. These services operate under Google's privacy terms. Website analytics are separate from your local Echologue journal and are not provided with journal content.
7. LEGAL BASES FOR PROCESSING
In Short: The legal basis depends on the feature and your location.
Where the GDPR, UK GDPR, or similar laws apply, we rely on the following legal bases:
- Consent: for transmitting journal, voice, or chat content to third-party AI services, and for optional device permissions or analytics where consent is required. You may withdraw consent at any time for future processing.
- Performance of a contract: to provide app features you request, manage purchases and subscriptions, restore entitlements, and provide support.
- Legitimate interests: to secure the Services, prevent abuse, diagnose errors, maintain reliability, and understand aggregate website usage, provided those interests are not overridden by your rights.
- Legal obligation: to comply with valid legal requirements, tax or accounting duties, and enforceable requests from competent authorities.
Withdrawing AI-processing consent does not affect the lawfulness of processing that occurred before withdrawal. It prevents new third-party AI transmissions after the setting takes effect.
8. HOW LONG INFORMATION IS KEPT
In Short: Local data remains until you delete it; third parties retain information according to their controls and legal obligations.
On-device data. Your local archive remains on your device until you edit or delete entries, use the in-app wipe function, clear app storage, or uninstall the app. Local recordings retained after a failed transcription remain on the device until you delete them or the app removes them as part of its normal workflow.
Our application infrastructure. We do not intentionally store a permanent copy of journal, voice, or chat content on our application servers. Requests are processed in transit. Limited security, operational, authentication, usage, and error records may be retained for the period reasonably necessary to operate and protect the Services.
AI service providers. OpenRouter-routed content is sent using zero-data-retention routing where configured and supported. Groq, OpenRouter, and endpoint providers may retain limited operational metadata and may process or retain information as required for security, abuse prevention, billing, legal compliance, or under their current published terms. We therefore do not claim that every technical record is destroyed instantly.
Purchase information. RevenueCat, Apple, and Google retain transaction and entitlement information according to their legal, accounting, fraud-prevention, and service requirements.
9. HOW WE PROTECT INFORMATION
We use platform application sandboxing, authenticated API access, Transport Layer Security (TLS), local database controls, limited-purpose data flows, and provider privacy controls designed to protect information in transit and at rest.
No security method is perfect. The security of locally stored information also depends on your device passcode, operating-system updates, backup settings, and physical control of the device. We encourage you to protect your device and keep its software updated.
10. INTERNATIONAL DATA TRANSFERS
Reformic OU is based in Estonia. Our service providers may process information in the European Economic Area, the United States, or other countries where they or their subprocessors operate. Those countries may have different data-protection laws.
Where required, we rely on lawful transfer mechanisms such as adequacy decisions, Standard Contractual Clauses, contractual safeguards, and provider data-protection commitments. You may contact us for more information about applicable safeguards.
11. CHILDREN'S PRIVACY
In Short: Echologue is not directed to children under 13, and local age requirements may be higher.
We do not knowingly solicit personal information from children under 13. In the European Economic Area, United Kingdom, or another region that requires a higher age for independent consent to online data processing, a child must have the authorization of a parent or legal guardian where required by law.
If you believe a child has transmitted personal information to our service providers without appropriate authorization, contact us. Local data can be deleted using the in-app wipe controls or by uninstalling the app.
12. YOUR PRIVACY RIGHTS AND CONTROLS
In Short: Most journal controls are available directly on your device.
Depending on your location, you may have rights to access, correct, delete, restrict, object to, or obtain a portable copy of personal information, and to withdraw consent or complain to a supervisory authority.
Because we do not maintain a user account or permanent cloud journal archive, we generally cannot identify or retrieve the local journal stored on your device. You can exercise direct control by reviewing, editing, exporting, or deleting information inside the app.
You can withdraw permission for third-party AI processing in Settings. This stops new AI transmissions but does not automatically erase your local entries or records that another provider is legally required to retain. For purchase or website information held by a provider, you may also need to exercise rights directly with Apple, Google, RevenueCat, Cloudflare, Groq, OpenRouter, or the relevant provider.
If you contact us about a privacy request, we may need enough information to verify the request and determine whether we hold any information associated with you. We will not ask you to send your journal archive merely to verify identity.
13. UNITED STATES PRIVACY DISCLOSURES
In Short: We do not sell personal information or share it for cross-context behavioral advertising.
The table below summarizes categories that may be processed in connection with the Services. Whether a category is processed depends on the features you use and the content you choose to enter.
| Category | Examples | Purpose and retention |
|---|---|---|
| Identifiers and technical information | IP address, request identifier, app-specific purchase identifier, device or app information | Service delivery, subscriptions, security, diagnostics, and fraud prevention; retained according to operational and legal requirements. |
| Commercial information | Product purchased, subscription and entitlement status, transaction status | Purchase processing and entitlement management; retained by Apple, Google, and RevenueCat according to their policies. |
| Audio and electronic information | Voice recordings, transcripts, journal text, and chat content | Local journaling, transcription, organization, embeddings, and grounded answers; local content remains until deleted, while remote processing follows provider controls. |
| Potentially sensitive personal information | Health, precise or approximate location within an entry, relationships, sexual life, beliefs, finances, or other sensitive journal content you choose to submit | Processed only to provide the journaling or AI feature you request; not used for advertising or model training authorized by us. |
| Inferences | Structured facts, feelings, themes, entities, memory summaries, and semantic vectors | Organize and retrieve your history; primarily stored locally until deleted. |
We have not sold personal information for money and do not share personal information for cross-context behavioral advertising. We do not knowingly sell or share the personal information of children.
14. OTHER REGIONAL RIGHTS
Residents of the European Economic Area, United Kingdom, Switzerland, Canada, Australia, New Zealand, Brazil, and other jurisdictions may have additional rights under local law. These can include the right to access, correct, erase, restrict, object, request portability, withdraw consent, and lodge a complaint with a data-protection authority.
For Estonia and the European Economic Area, you may contact the Estonian Data Protection Inspectorate or the supervisory authority in your country of residence. You may also contact us using the details below.
15. DO-NOT-TRACK SIGNALS
There is no universally accepted standard for responding to browser Do-Not-Track signals. Our mobile app does not use journal content for targeted advertising or cross-app tracking. Our website may use basic analytics as described above and will follow applicable consent requirements.
16. CHANGES TO THIS NOTICE
We may update this Privacy Notice to reflect changes to the Services, providers, data flows, or legal requirements. The revised date appears at the top. If a change materially affects how personal journal, voice, or chat content is shared, we may provide an in-app notice and request renewed consent where required.
17. HOW TO CONTACT US
If you have questions, comments, or a privacy request, contact:
Reformic OU
Privacy Contact
139e/2 Pärnu mnt.
Tallinn, Harju maakond 11317
Estonia
Email: reformic.est@gmail.com
18. HOW TO REVIEW, EXPORT, OR DELETE YOUR DATA
You can manage your local Echologue archive directly in the app:
- Review and edit: Open an entry or remembered chat and use the available editing or deletion controls.
- Export: Export your journal in supported portable formats, including JSON or CSV.
- Withdraw AI permission: Open Settings and disable third-party AI processing. This blocks new AI transmissions.
- Complete local deletion: Use "Wipe Data" in Settings to erase local journal text, structured fields, vectors, remembered chats, memory state, and related app records.
- Uninstall: Uninstalling the app generally removes its local application data, subject to your operating system's backup and restoration settings.
Because we do not maintain a permanent server-side journal archive, there is normally no cloud journal account for us to delete on your behalf. Transaction, security, or operational records held by third-party providers may remain for their required retention periods.