ChefAI documentation

AI-powered recipe generator, weekly meal planner and calorie tracker built with Flutter. Complete source code for CodeCanyon buyers.

ChefAI is a full-featured Flutter mobile app that lets users generate AI-powered recipes from ingredients they already have, plan weekly meals, track calories and macros, and follow interactive step-by-step cooking mode — all powered by OpenAI or Google Gemini.

This package includes the complete Flutter source code, Firebase configuration, and all assets needed to publish on the App Store and Google Play under your own brand.

Key features

Tech stack

• Flutter 3.24
• Dart 3.5
• Riverpod 2
• Firebase
• OpenAI GPT-4o
• Gemini 3.5 Flash
• iOS and Android

1. 01

   Extract the source code

   Unzip the downloaded package. You will find the Flutter project inside the chef-ai/ folder.
2. 02

   Install Flutter dependencies

   Open a terminal in the project root and run:

   terminalCopy

   flutter pub get

3. 03

   Generate code (freezed models and Riverpod)

   This step is required whenever models change. Run:

   terminalCopy

   flutter pub run build_runner build --delete-conflicting-outputs

4. 04

   Install iOS pods

   Required for iOS builds only:

   terminalCopy

   cd ios && pod install && cd ..

5. 05

   Complete Firebase setup (see next section)

   The app will not run without Firebase configuration files.

ChefAI uses Firebase for authentication, Firestore database, Remote Config (API keys), push notifications, and analytics.

Create a Firebase project

1. 01

   Go to console.firebase.google.com

   Sign in with a Google account and click Add project.
2. 02

   Register Android app

   Package name: com.devsnack.chefai (change to your own ID). Download google-services.json and place it in android/app/.
3. 03

   Register iOS app

   Bundle ID: com.devsnack.chefai (change to your own ID). Download GoogleService-Info.plist and place it in ios/Runner/.

Enable Firebase services

In the Firebase Console, enable the following services under your project:

Apply Firestore security rules

In Firestore → Rules tab, paste the following:

// Firestore Security Rules — paste in Firebase Console → Firestore → Rules
rules_version = '2';
service cloud.firestore {
  match /databases/{database}/documents {
    match /users/{userId} {
      allow read, write: if request.auth.uid == userId;
    }
    match /recipes/{recipeId} {
      allow read: if request.auth != null &&
        (resource.data.userId == request.auth.uid || resource.data.userId == null);
      allow create: if request.auth != null &&
        request.resource.data.userId == request.auth.uid;
      allow update, delete: if request.auth != null &&
        resource.data.userId == request.auth.uid;
    }
    match /mealPlans/{userId}/weeks/{weekId} {
      allow read, write: if request.auth.uid == userId;
    }
    match /shoppingLists/{userId}/weeks/{weekId} {
      allow read, write: if request.auth.uid == userId;
    }
    match /nutritionLogs/{userId}/days/{date} {
      allow read, write: if request.auth.uid == userId;
    }
  }
}

All API keys and feature flags are stored in Firebase Remote Config — never hardcoded in the app. This allows you to update keys without releasing a new app version.

Add Remote Config parameters

In Firebase Console → Remote Config → Add parameter:

Getting API keys

• OpenAI: visit platform.openai.com → API Keys → Create new secret key. Make sure your account has Chat Completions access (requires a paid plan).
• Google Gemini: visit aistudio.google.com → API Keys → Create API key. Gemini 3.5 Flash / Flash-Lite are available on the free tier with rate limits.

ChefAI uses RevenueCat to manage premium subscriptions across iOS and Android with a single integration.

1. 01

   Create a RevenueCat account

   Go to app.revenuecat.com and create a new project named ChefAI.
2. 02

   Add iOS and Android apps

   Add your bundle ID for iOS and package name for Android. Copy the Public API Key for each platform.
3. 03

   Create products in App Store Connect / Google Play

   Create subscription products (e.g. chefai_premium_monthly, chefai_premium_yearly). Import them into RevenueCat.
4. 04

   Update the app

   Open lib/data/services/revenue_cat_service.dart and replace the placeholder API keys with your RevenueCat public API keys.

   revenue_cat_service.dartCopy

   // lib/data/services/revenue_cat_service.dart
   await Purchases.configure(
     PurchasesConfiguration('your_revenuecat_public_key')
   );

1. 01

   Create an AdMob account

   Go to admob.google.com. Create your account and add an app for both iOS and Android.
2. 02

   Create ad units

   Create a Banner ad unit and an Interstitial ad unit. Copy the Ad Unit IDs.
3. 03

   Add App IDs to native config

   The AdMob App ID (not ad unit ID) must be added to native files:

   Android — android/app/src/main/AndroidManifest.xml

   AndroidManifest.xmlCopy

   <meta-data
       android:name="com.google.android.gms.ads.APPLICATION_ID"
       android:value="ca-app-pub-XXXXXXXXXXXXXXXX~XXXXXXXXXX"/>

   iOS — ios/Runner/Info.plist

   Info.plistCopy

   <key>GADApplicationIdentifier</key>
   <string>ca-app-pub-XXXXXXXXXXXXXXXX~XXXXXXXXXX</string>

4. 04

   Set ad unit IDs in Remote Config

   Add admob_banner_id and admob_interstitial_id in Firebase Remote Config (see Remote Config section above).

All colors are centralized in lib/core/constants/app_colors.dart. Never hardcode colors in widgets — always use AppColors constants.

Change the primary color

The default accent is Saffron Orange (#F97316). To rebrand with a different color, update these two values in app_colors.dart:

static const primary      = Color(0xFFF97316);  // ← change this
static const primaryLight = Color(0xFFFB923C);  // ← and this (lighter shade)
static const primaryTint  = Color(0xFFFED7AA);  // ← background tint

Change app icon

Replace the icon assets:

• Android: Replace files in android/app/src/main/res/mipmap-*/
• iOS: Replace assets in ios/Runner/Assets.xcassets/AppIcon.appiconset/

Or use the flutter_launcher_icons package for automated generation from a single source image.

Change app name

Change bundle / package ID

ChefAI uses Montserrat loaded via the google_fonts package. No font files need to be bundled — they are downloaded automatically on first launch and then cached.

Typography scale

To change the font family, replace GoogleFonts.montserrat() with any other google_fonts font in lib/core/theme/app_theme.dart.

Strings and copy

All user-facing text is stored in lib/core/constants/app_strings.dart. To translate or rebrand the app copy, edit only this file — do not change strings directly in widgets.

The app is English only in v1. Multi-language support is planned for v2.

Folder structure

lib/
├── core/
│   ├── constants/      ← colors, strings, sizes, prompts
│   ├── enums/          ← AIProvider, MealType, DietaryPreference…
│   ├── theme/          ← light & dark themes
│   ├── router/         ← GoRouter routes + auth redirect
│   └── utils/          ← NutritionCalculator, ServingScaler…
│
├── data/
│   ├── models/         ← freezed models (Recipe, MealPlan…)
│   ├── repositories/   ← Firestore read/write logic
│   └── services/
│       ├── ai/         ← RecipeAIService + OpenAI + Gemini + Factory
│       ├── remote_config_service.dart
│       ├── revenue_cat_service.dart
│       └── admob_service.dart
│
├── presentation/
│   ├── providers/      ← Riverpod providers (auth, recipe, meal plan…)
│   ├── screens/        ← one folder per feature / screen
│   └── widgets/        ← common, recipe, nutrition, meal_plan, cooking
│
└── main.dart

State management

ChefAI uses Riverpod 2 with code generation. All business logic lives in lib/presentation/providers/. Key providers:

Navigation

ChefAI uses GoRouter with a shell route for the bottom navigation bar. Auth-guard redirects unauthenticated users to /login and unboarded users to /onboarding.

// Always navigate with GoRouter — never use Navigator.push
context.go('/home');           // replace stack
context.push('/recipe/123');   // push on stack
context.pop();                 // go back

AI calls go through an abstract interface RecipeAIService. Screens and providers never know which AI provider is active — the factory picks the right implementation based on Remote Config.

Switching AI provider

Change the default_ai_provider value in Firebase Remote Config to openai or gemini. The change takes effect on the next app fetch (within 1 hour) — no app update needed.

AI methods

Adding a new AI provider

1. 01
   Create a new class implementing RecipeAIService
2. 02
   Add the provider to AIProvider enum
3. 03
   Add a case to AIServiceFactory.create()
4. 04
   Add the API key to Remote Config

All models use freezed for immutability and JSON serialization. Never write models by hand.

After any model change, always regenerate:

flutter pub run build_runner build --delete-conflicting-outputs

Limits are enforced in providers, not screens. All values come from Remote Config so you can adjust them without an app update.

Build commands

# Install dependencies
flutter pub get

# Regenerate code after model changes
flutter pub run build_runner build --delete-conflicting-outputs

# Format code
dart format lib/

# Static analysis (must be zero issues)
flutter analyze

# Run in debug mode
flutter run

# Build release APK
flutter build apk --release

# Build release App Bundle (Google Play)
flutter build appbundle --release

# Build release iOS (requires macOS + Xcode)
flutter build ios --release

Android release

1. 01

   Generate a keystore

   terminalCopy

   keytool -genkey -v -keystore ~/upload-keystore.jks -alias upload -keyalg RSA -keysize 2048 -validity 10000

2. 02

   Configure signing

   Create android/key.properties with your keystore path, alias, and passwords. Reference it in android/app/build.gradle.
3. 03

   Update version

   In pubspec.yaml, set version: 1.0.0+1 (name+build number).
4. 04

   Build App Bundle

   flutter build appbundle --release → upload build/app/outputs/bundle/release/app-release.aab to Google Play.

iOS release

1. 01

   Open in Xcode

   Open ios/Runner.xcworkspace (not .xcodeproj) in Xcode.
2. 02

   Set team and bundle ID

   In Runner → Signing and Capabilities, select your Apple Developer Team and set the Bundle Identifier to your own ID.
3. 03

   Add Sign in with Apple capability

   Click + Capability → add Sign In with Apple.
4. 04

   Archive and upload

   Product → Archive → Distribute App → App Store Connect → Upload.

Pre-launch checklist

• Change bundle ID / package name from com.devsnack.chefai to your own
• Place google-services.json in android/app/
• Place GoogleService-Info.plist in ios/Runner/
• Enable Firebase Auth (Email, Google, Apple)
• Apply Firestore security rules
• Add all Remote Config parameters (API keys, limits, AdMob IDs)
• Add AdMob App ID to AndroidManifest.xml and Info.plist
• Configure RevenueCat and update API key in revenue_cat_service.dart
• Update app name in AndroidManifest.xml and Info.plist
• Replace app icon assets
• Run flutter analyze — zero issues
• Test on a real iOS device and Android device
• Test purchase flow in sandbox / test environment
• Test AdMob ads are loading
• Verify dark mode on all screens

Every version published so far. Updates are free for the life of the item.

Version 1.2.0 · August 2026

• Smart shopping list — build a grocery list from the weekly meal plan. Duplicate ingredients are merged across recipes, grouped by supermarket aisle, with tick-off state saved to Firestore, manual items, and plain-text sharing.
• AI model IDs moved to Remote Config (openai_model, gemini_model, and their _fast variants) — change models without an app update when a provider retires one.
• Default Gemini models updated to the current generation.
• Editable profile (/profile/edit) — diet, allergies, cuisines, calorie and macro goals, household size, cooking skill, and measurement units, with a BMR/TDEE calorie calculator.
• Paywall now loads real RevenueCat offerings and purchases the selected package.
• Request timeouts and detailed error logging added to both AI services; token budgets now scale with the requested recipe count.
• Fixes: cooking timer notifications now initialize and fire; recipe ratings and cook counts persist for unsaved recipes; the logging streak increments; fiber is tracked per logged meal; scaled servings carry into cooking mode.

Version 1.1.0 · July 2026

• Manual meal logging to any meal section.
• AI nutrition lookup surfaced in the tracker.
• Shared AI service provider used by both recipe generation and nutrition lookup.

Version 1.0.0 · March 2026

• Initial release — all screens, dual AI providers, RevenueCat + AdMob monetization, full dark mode, step-by-step cooking mode with wakelock and timers.

We’re happy to help with installation, configuration, and customization questions. Typical response time is within 24 hours on business days.

What support covers

• Six months of support from purchase, extendable at checkout
• Support covers bugs in the template and questions about how it is put together
• It does not cover custom feature work, third-party API changes or store review outcomes
• The Regular licence covers one free end product. Charging users for the app itself needs the Extended licence