Privacy Policy

Version
2026-09-04
Effective date
2026-09-04

TypeCal is provided by TypeGen Studio, an independent software business based in India.

This Privacy Policy explains what information TypeCal processes, why it is processed, who receives it, how retention and deletion work, and the choices available to you.

Privacy at a Glance

  • TypeCal processes account, nutrition, wellness, and feature information to provide the services you request.
  • Food, recipe, image, audio, and other automated results are estimates and should be reviewed.
  • Apple Health access is optional. Apple Health and other health and wellness information is not used for advertising.
  • TypeCal does not sell or rent personal information.
  • TypeCal does not use your content to train its own general-purpose AI models.
  • MCP connections begin with approved read access. Write permissions are separate and are off by default for each connection.
  • You can delete supported items, delete your account, manage device permissions, and contact support@typegenstudio.com about a privacy request or grievance.

Scope

This Policy applies to the TypeCal mobile app and its iPhone and iPad interfaces, backend services, website and waitlist, browser authentication and OAuth pages, widgets and Live Activities, support and feedback channels, and the optional TypeCal MCP service. A third-party app or AI client that you connect processes information it receives under its own terms and privacy policy.

Information We Collect

Account, identity, and profile. We process authentication identifiers, email address, sign-in provider, display name, first and last name if supplied, avatar, security state, and account timestamps. We also process onboarding and settings information such as age, sex, height, weight, activity level, goals, preferred pace, units, dietary preferences, allergies, notification choices, and how you heard about TypeCal.

Food, nutrition, hydration, and progress. We process meal descriptions, ingredients, calories, nutrients, serving details, dates and times, hydration, body weight, targets, activity summaries, edits, saved meals, streaks, achievements, and progress data.

Recipes, pantry, and grocery content. We process recipes, ingredients, steps, source links, thumbnails, saved collections, pantry items, grocery items, store preferences, and changes you make to that content.

Photos, barcodes, labels, voice, and imports. When you use these features, TypeCal may process photos and images, barcodes, nutrition labels, typed or pasted text, voice audio and transcripts, recipe URLs and webpage data, social-media video or audio, extracted frames, optical-character-recognition results, remote thumbnails, temporary working files, and structured or derived recipe and nutrition content. Media attached to saved meals and recipes may be stored in account-scoped storage.

Apple Health. If you grant permission, the TypeCal app may read selected Apple Health categories, including dietary energy and nutrients, body weight, steps, and active energy. The app may write nutrition entries to Apple Health when you direct it to save them. Limited activity summaries may be sent to the TypeCal backend for progress and rewards. You control access through Apple Health or iOS Settings.

Subscriptions and purchases. Apple and RevenueCat process subscription purchases and trials. TypeCal receives entitlement, product, trial, renewal, and provider identifiers needed to manage access. TypeCal does not receive full payment-card details.

Website, waitlist, notifications, widgets, and Live Activities. If you join the waitlist, we process your email address, signup source, limited user-agent, page, referrer, campaign, session, visitor, time-on-page, delivery, and provider-message information. We process device notification tokens, preferences, delivery and interaction events, and selected notification moments. Widgets and Live Activities can display nutrition, hydration, and progress information on the Home Screen or Lock Screen, where someone with access to your device may see it.

Product analytics, session replay, and diagnostics. TypeCal uses PostHog for product analytics and samples approximately 25% of in-app sessions for replay to find and fix product problems. In replay, text inputs and images are masked. The app's replay configuration does not capture network telemetry or application logs. We also process device and app details, feature-use events, request metadata, and error, performance, security, and troubleshooting information.

TypeCal's backend AI observability can be configured to send prompt text and generated output to PostHog. The deployed state of that switch has not been verified for this Policy, so we do not state that AI content is always sent or never sent. This optional backend capture is separate from masked session replay.

MCP, support, and feedback. For an MCP connection, we process the authorized client, permission, request, response-size, quota, timing, outcome, and security metadata described below. If you contact us, submit private feedback, or post on the public feedback board, we process the message, display name, attachments you choose to provide, and information needed to publish, investigate, or reply.

How We Use Information

We use information to:

  • create, authenticate, maintain, and secure accounts;
  • provide, synchronize, and personalize food logging, nutrition estimates, hydration, recipes, pantry, grocery, progress, rewards, and other TypeCal features;
  • process food, images, audio, labels, webpages, and imported content into estimates;
  • operate Apple Health, subscription, notification, widget, and Live Activity features you enable;
  • send requested waitlist or service communications and honor unsubscribe choices;
  • provide and secure OAuth and MCP connections you authorize;
  • measure use, investigate errors, enforce quotas, prevent abuse, and improve reliability, accuracy, and safety;
  • respond to support, feedback, privacy, and grievance requests; and
  • comply with applicable law and enforce the Terms of Use.

AI and Automated Processing

TypeCal uses automated systems, nutrition databases, computer vision, speech recognition, and Google Gemini to produce food, ingredient, portion, nutrient, and recipe estimates. TypeCal may also use Groq when audio transcription is enabled. The relevant input, such as text, an image, audio, a transcript, or imported recipe content, may be sent to the applicable provider. Outputs can be inaccurate or incomplete, so review and edit them before relying on them. TypeCal does not use automated results to provide medical advice or make legal or similarly significant decisions about you.

TypeCal does not use your content to train its own general-purpose AI models. This statement describes TypeCal's own use; external AI providers control their processing under the provider arrangement and settings that apply to the request.

MCP and Connected Clients

If you connect an AI client or another compatible application, TypeCal discloses the account and data categories shown during authorization. Read tools can return authorized account, profile and target, daily summary, food-log, nutrition-progress, hydration-progress, saved-meal, recipe, and grocery information.

Write permissions are separate from the initial read connection and are off by default for each connection. The six supported actions are available only when MCP writes are enabled and separately authorized. Three separate permissions govern grocery changes, recipe changes, and log changes for food and hydration. Logging a recipe with nutrition also requires recipe permission because it updates the recipe's stored nutrition. The six supported write actions are:

  • Add Grocery Items
  • Update Grocery Item
  • Add Recipe Ingredients to Grocery List
  • Create Recipe
  • Log a Recipe as Eaten
  • Log Water

An authorized client acts using the permissions you grant, but AI clients can misunderstand requests or create errors or duplicates. Review resulting changes in TypeCal. No current MCP tool deletes TypeCal data. No current MCP tool writes to Apple Health.

TypeCal records limited MCP security and disclosure metadata, including account ID, client identifier, request ID, tool name, success or failure, error code, duration, response size, quota use, and backend version. MCP audit records are designed not to contain tool arguments, tool results, food or recipe content, prompts, or access tokens.

Disconnecting or revoking a client prevents new TypeCal access after revocation completes. It cannot retrieve or delete copies already received or retained by the third-party client or its provider. Use that provider's controls for information held by it.

Sharing and Service Providers

TypeCal does not sell or rent personal information. Apple Health and other health and wellness information is not used for advertising. We disclose information only as needed to operate requested features, secure TypeCal, process purchases, respond to you, comply with law, protect rights and safety, or complete a transaction involving the service subject to appropriate protections.

The core providers used by TypeCal are:

  • Supabase for PostgreSQL database, authentication, and object storage services;
  • Render for backend, worker, media-processing, and operational-log hosting;
  • Google Gemini for AI processing;
  • PostHog for product analytics, sampled and masked session replay, and configurable backend AI observability;
  • RevenueCat for subscription and entitlement management;
  • Apple for sign-in, the App Store, Apple Health, notifications, Siri or Shortcuts, widgets, and Live Activities; and
  • Google for sign-in.

Configured or feature-dependent integrations include:

  • Groq for audio transcription when audio transcription is enabled; and
  • Resend for waitlist contacts and email delivery when waitlist email delivery is enabled.

An AI client or compatible application you choose to connect through MCP also receives the information you authorize. We may disclose information to legal, safety, or compliance recipients when required by law or needed to protect rights and safety, and to a successor in a transaction involving TypeCal subject to appropriate protections.

International Processing

TypeCal is operated from India, but information is not necessarily stored or processed in India. TypeCal and its providers may process information in the United States and other countries where they operate. Those countries may have privacy laws different from those where you live. Where required, we use appropriate safeguards for international transfers. We do not represent that any provider uses a particular processing region unless that setting has been verified for TypeCal.

Retention and Deletion

We keep information only while it is needed to provide TypeCal, maintain an account or requested feature, secure and troubleshoot the service, process transactions, respond to disputes, meet legal obligations, or support the provider-controlled purposes described in this Policy. Where no verified fixed deletion period is stated below, retention follows those purposes, necessity, applicable law, and the verified settings of the provider that holds the information.

For recipe-import processing, protected records and uploaded objects are assigned an expiry window:

  • encrypted protected recipe-resolver evidence is assigned an expiry of up to 7 days; related non-content integrity and operational metadata may remain while needed;
  • an uploaded resolver-media object is assigned an expiry after 1 hour if it is not referenced and after 24 hours once referenced; and
  • temporary remote social-media acquisition files are created only when that import feature runs and are removed after processing or during later cleanup.

The recipe-import cleanup process removes expired protected content and media when it runs. If cleanup is not active when an expiry is reached, removal can occur later than the assigned expiry.

You can edit or delete supported logs, recipes, media, and other items. The account-deletion workflow revokes account access, deletes many directly account-owned database records, and queues known account media for deletion and retry. Short-lived deletion-operation records remain until queued work is resolved and deletion is verified. One-way SHA-256 fingerprints of retired media locators may remain for deletion integrity; they do not contain the raw media locator. Deleting a TypeCal account does not itself cancel an App Store subscription.

Some current systems do not follow account deletion automatically or do not have a verified fixed expiry:

  • an account-scoped nutrition parse cache may remain after account deletion because it is outside the main account-deletion path and does not currently have automatic expiry;
  • AI cost events and image-parse attempt records may remain after their direct user ID is removed, with request, model, usage, timing, outcome, or diagnostic metadata kept while needed for cost control, reliability, security, and accounting;
  • public feedback-board posts and comments may retain the original display name and authored text after the user ID is removed, and public roadmap text derived from private feedback may remain after its source link is removed;
  • a shared branded-nutrition verification cache can retain a normalized query and derived nutrition, validation, and source information without an automatic deletion period;
  • rejected branded-nutrition verification logs can include the submitted query text and a source URL; Render holds application logs under the settings applicable to TypeCal's service;
  • deleting an account or resetting the app's analytics identity does not by itself confirm remote deletion of PostHog events or session replay;
  • deleting a TypeCal account does not automatically delete RevenueCat records or remotely revoke every Apple or Google identity grant; and
  • provider-controlled analytics, other logs, AI request data, waitlist delivery history, subscription and transaction records, identity records, and database or storage recovery copies remain under their applicable purposes, legal requirements, and verified provider settings. Deleted information may persist temporarily in recovery copies until those copies expire.

MCP audit and quota metadata, notification engagement events, waitlist contacts, application and security logs, and other operational records are retained only while needed for their stated purposes or under verified provider settings. TypeCal does not claim a fixed period for those records unless the period is active and verified. If a documented legal or security obligation requires us to preserve a scoped record, we may keep that record until the obligation ends and then resume the applicable deletion process.

Disconnecting an MCP client does not delete information already received by that client. Apple Health data on your device remains controlled through Apple Health and iOS; TypeCal does not control all HealthKit retention.

Security

TypeCal uses administrative, technical, and organizational measures designed to protect information, including authenticated account-scoped access, private media storage, signed media URLs, rate limits, delegated MCP permission controls, and deletion retry workflows. Access to personal information is limited according to role and purpose. No system is completely secure, and we cannot guarantee absolute security.

Your Choices and Rights

Depending on where you live, you may have rights to request access, correction, export, deletion, restriction or objection, withdrawal of consent, or information about processing. You may also have a right to raise a grievance or appeal a response. We will not discriminate against you for exercising an applicable privacy right.

You can:

  • edit or delete supported TypeCal records and delete your account in the app;
  • manage Apple Health, camera, microphone, notification, Siri or Shortcuts, and widget permissions in the applicable iOS settings;
  • turn off Lock Screen progress in TypeCal settings;
  • manage or cancel a subscription through Apple;
  • disconnect MCP clients and revoke individual write categories through the connection manager;
  • unsubscribe using the private link in a waitlist email; and
  • contact support@typegenstudio.com to request access, correction, export, deletion, consent withdrawal, or grievance review.

We may need to verify your identity before acting on a request. Some rights and exceptions depend on your location and applicable law.

Children's Privacy

TypeCal is intended for users age 13 and older. If you are at least 13 but under the age of majority where you live, by using TypeCal you represent that you have permission from a parent or legal guardian. TypeCal does not store a separate guardian-consent record or independently verify that permission. If you believe a child is using TypeCal contrary to this rule, contact us so we can investigate and take appropriate action.

Changes to This Policy

We may update this Policy as TypeCal changes. We will update the version and effective date. For material changes, TypeCal will provide in-app notice and acknowledgement. We will also send email when legally required or when a change materially affects your rights.

Contact

TypeGen Studio is responsible for monitoring and responding to privacy and grievance requests for TypeCal. TypeCal is provided by TypeGen Studio, an independent software business based in India. Email support@typegenstudio.com.