MyFitnessPal Premium

Build me a private meal and activity journal for iOS and Android, as a narrow substitute for MyFitnessPal Premium. Requirements:

- Use React Native with Expo, TypeScript, and device-local SQLite. Model user-chosen meal entries, activity entries, daily notes, optional routines, and imported health records with stable IDs.
- Let me record a meal or activity with date, units, and a free-text note. Corrections update the daily summary and remain visible in edit history.
- Show a weekly pattern view using neutral language. Offer moderate routines and rest days without weight-loss targets, restrictions, scores, or medical conclusions.
- HealthKit or Health Connect import is optional and permission-scoped by data type. The app must work fully with permission denied and deduplicate imports by platform record ID.
- Export a selected date range as JSON/CSV and offer full local deletion. No accounts, analytics, cloud upload, or food database claims.
- Acceptance: add two manual entries, correct one, import one duplicate health record, and verify the day total includes each entry once; deny health permission and confirm manual logging/export still works.
- README covers local backup, permission removal, and the non-clinical scope.

# AGENTS.md — Build guide for MyFitnessPal Premium

## Project scope
Log user-chosen meals and activity, plan moderate routines, and show neutral summaries and trends without diagnosing, shaming, or enforcing restrictive goals.

Catalogue verdict: no. A consolation build is possible, but the paid product's decisive value sits outside a solo rebuild. For MyFitnessPal Premium, log user-chosen meals and activity with neutral summaries and no restrictive coaching. The hard boundary is huge food database, barcode data, device integrations, models, and mobile polish, plus hardware data, content, coaching, and trust.
Use the implementation prompt below to define the deliverable. Complete each phase's acceptance checks before extending the scope.

## Working agreement
- Inspect the repository and its existing instructions before choosing paths, dependencies or commands. Keep one coherent stack and explain changes to the proposed architecture.
- Plan a vertical slice that accepts a real input and produces the useful output described below. Persist only the state the prompt calls for; respect memory-only and upstream-managed workflows. Use fixtures only when they are clearly labelled.
- After scaffolding, document the actual install, development, check and build commands in README and keep them synchronized with the package or project manifest. Do not report commands as successful unless they ran.
- Work in small steps. At handoff, list implemented flows, checks actually performed, remaining blockers, and any credentials or provider setup the owner must supply.
- Do not publish, spend money, contact customers, delete source data or run irreversible migrations without the project owner's authorization.

## Prerequisites
- iOS or Android device
- optional HealthKit or Health Connect permission
- local storage

## Stack and architecture
- React Native with Expo and TypeScript for native mobile screens
- Device-local SQLite for meal, activity, note, routine, and optional imported-health records

## Security and data integrity
- Request health-data permissions per type and remain usable when denied
- Deduplicate platform imports by source record ID and keep user data on device
- Support full deletion and dated export without medical interpretation

## Agent implementation rules
- Separate manually entered records from optional health-platform imports
- Recompute daily summaries after edits or deletion
- Test duplicate imports and permission denial with fixture records

## Optional agent skills and references
No installable skill is listed for this guide; use the project rules and the stack documentation.

Read the linked SKILL.md and its dependencies before adding a skill. Select only the skills matching this project's runtime and task; their documentation does not supply API access, credentials or approval to perform external actions. Pin the reviewed revision where the tool supports it. Follow the chosen agent's documented project-level installation mechanism.

## Distribution ideas
These are optional planning notes. Obtain the owner's approval before publishing or contacting anyone.
- Demonstrate neutral weekly journaling with synthetic entries and permission denial
- Explain the absence of the paid food catalog, barcode inventory, coach network, and clinical validation

## Engineering roadmap
1. Phase 1: Define SQLite tables for manual entries, routines, notes, and imported records with source IDs; seed a two-day fixture.
2. Phase 2: Build native entry/edit screens and recompute daily totals from current records rather than cached counts.
3. Phase 3: Add opt-in, per-type health imports; denial leaves manual journaling and export functional.
4. Phase 4: Render neutral weekly summaries and JSON/CSV export, plus complete local deletion.
5. Phase 5: Correct an entry, replay a duplicate platform record, and verify totals; document non-clinical limits and backup steps.

## Paid-product capabilities outside this build
- huge food database, barcode data, device integrations, models, and mobile polish
- wearable hardware data
- coach network
- clinical validation
- large content library and social network

## Implementation prompt
Build me a private meal and activity journal for iOS and Android, as a narrow substitute for MyFitnessPal Premium. Requirements:

- Use React Native with Expo, TypeScript, and device-local SQLite. Model user-chosen meal entries, activity entries, daily notes, optional routines, and imported health records with stable IDs.
- Let me record a meal or activity with date, units, and a free-text note. Corrections update the daily summary and remain visible in edit history.
- Show a weekly pattern view using neutral language. Offer moderate routines and rest days without weight-loss targets, restrictions, scores, or medical conclusions.
- HealthKit or Health Connect import is optional and permission-scoped by data type. The app must work fully with permission denied and deduplicate imports by platform record ID.
- Export a selected date range as JSON/CSV and offer full local deletion. No accounts, analytics, cloud upload, or food database claims.
- Acceptance: add two manual entries, correct one, import one duplicate health record, and verify the day total includes each entry once; deny health permission and confirm manual logging/export still works.
- README covers local backup, permission removal, and the non-clinical scope.

## Completion evidence
Demonstrate the prompt's acceptance scenarios against the scoped workflow. Include setup from a clean checkout and failure recovery. Check persistence across restart and export/restore only for the state the prompt says to store; for memory-only tools, confirm that temporary content is discarded as specified. Record actual results and remaining limitations. A detailed plan alone does not establish a working replacement.