Privacy Policy
LinguaMind is built by Mick Johnson ("we" / "us"). This policy explains what we collect, why, and what we don't. Plain English; no dark patterns.
What we collect
- Text you paste. When you feed an email, journal entry, article, or transcript into LinguaMind, we send that text to our backend on Cloudflare to extract vocabulary via Anthropic's Claude model. The text and the extracted vocabulary are stored in our Cloudflare D1 database, scoped to your account, so your course persists across sessions and devices.
- Your Apple identifier (iOS, when launched). Once the iOS app ships, Sign in with Apple will give us a stable, opaque identifier (the
subclaim) to link your account across devices. We will not request your name or email unless you grant them at sign-in. The identifier will be stored in your iOS Keychain and in our Cloudflare D1 database. Not collected today (browser-only). - Subscription state (iOS, when launched). Once Pro ships on iOS, Apple's StoreKit will tell the app whether you hold an active subscription. We will not see your payment instrument.
- Diagnostic crash reports (iOS, when launched). Once iOS ships, Apple will send us anonymous crash logs through the standard iOS diagnostics opt-in. Optionally, anonymized error events flow through Sentry — only if you have not disabled iOS diagnostic sharing.
What we don't collect
- No third-party tracking. No advertising SDKs. No IDFA. No fingerprinting.
- No microphone, camera, location, contacts, calendar, or photos access. LinguaMind never requests these permissions.
- No purchase history. (Once iOS launches, only Apple's StoreKit subscription status.)
- No selling of any data, ever.
Third-party recipients
- Cloudflare — hosts our backend (Pages + Workers + D1 + KV). Your data is processed and stored on Cloudflare's infrastructure. See Cloudflare's privacy policy.
- Anthropic — we call Claude (Haiku model) via the Anthropic API for vocabulary extraction and AI tutor sessions. Your pasted text passes through Anthropic's API as part of the extraction step. Anthropic does not train on this data per their API terms.
- Apple — once the iOS app launches, for Sign in with Apple and App Store transactions only. Not in use today (browser-only).
- Sentry — optional anonymized error telemetry. No personally-identifying data is sent.
Your rights
- Delete your data. Settings → Sign Out → Delete Account. We remove your vocabulary graph and source library from our database within 30 days. (Once iOS launches, your Apple identifier is removed on the same window.)
- Export your data. Settings → Export my data (JSON). Produces a JSON file with your vocabulary, sources, and review history.
- Stay local. Tap "Delete all local data" in Settings to clear this device's cache without touching the cloud copy.
Children
LinguaMind is not directed at children under 13. We do not knowingly collect data from children under 13.
Changes
If this policy changes, the updated version is hosted at this URL with an updated effective date. Material changes will surface a notice inside the app on next launch.