Recipe & Meal Planning

Part of the worlds-biggest-software-project initiative.

An AI-native, open-source meal planner that unifies recipes, pantry inventory, nutrition tracking, and grocery lists into a single coherent workflow.

Recipe & Meal Planning generates personalised weekly meal plans from dietary preferences, household profiles, and what is already in the pantry, then turns those plans into smart grocery lists and nutritional summaries. It is built for households who want healthy, affordable meals without juggling separate apps for recipes, shopping, and nutrition.

Why Recipe & Meal Planning?

Existing tools operate in silos: recipe apps, grocery-list tools, and nutrition trackers force users to duplicate effort across platforms.
Incumbents like Eat This Much offer macro-targeted automation but have a dated UI, limited recipe customisation, and no pantry inventory awareness.
MyFitnessPal locks meal planning and barcode scanning behind Premium+ paywalls, and its meal plans are not dynamically adaptive.
Mealime and Plan to Eat have no AI plan generation, while Plan to Eat also lacks nutrition data, pantry management, and grocery delivery.
Yummly's roadmap is uncertain following its acquisition by Whirlpool, and no open-source meal planner with comparable feature depth currently exists.

Key Features

AI Planning & Recipes

AI meal plan generator that builds weekly plans from dietary preferences, calorie targets, household size, and pantry inventory
Recipe library with URL import, user-submitted recipes, and version history
Social media recipe extraction from sources such as TikTok, Instagram, and YouTube
Natural-language adjustments to plans (e.g. budget-friendly, ingredient swaps)

Grocery & Pantry

Smart grocery list auto-aggregated from the weekly plan, de-duplicated and organised by supermarket aisle
Pantry manager with inventory tracking and expiry reminders to reduce food waste
Grocery delivery integration via Instacart, DoorDash, or retailer APIs
Barcode scanning for pantry item entry and grocery receipt parsing

Nutrition & Dietary Fit

Macro and micronutrient breakdown per meal and per day with goal tracking
Dietary filter suite covering vegan, gluten-free, keto, and allergy-aware filtering applied across all plan generation
Nutritional data backed by USDA FoodData Central
Family/household mode with multiple preference profiles resolved into a single coherent plan

AI-Native Advantage

LLM-powered plan generation lets the system satisfy nutritional constraints, household preferences, and pantry inventory simultaneously — something incumbents do not do dynamically. AI ingredient substitution, recipe extraction from unstructured sources (videos, images, social posts), and natural-language plan adjustments turn meal planning from a manual browsing exercise into a conversational workflow. Pantry-aware planning closes the loop between what is on hand and what gets cooked, directly reducing waste.

Tech Stack & Deployment

The system is designed around an LLM-driven planning core with constraint satisfaction over nutrition, preferences, and pantry state, fine-tuned on recipe and nutritional datasets. Nutrition data uses the public USDA FoodData Central; open standards such as Schema.org Recipe markup and Cooklang are available for interoperability. Real-time integrations target the Instacart and Kroger APIs for live pricing and delivery, with offline recipe and shopping list functionality for in-store use. Cross-platform delivery covers web, iOS, and Android.

Market Context

The AI-driven meal planning market grew from $0.83 billion in 2025 to $1.03 billion in 2026 (24.6% CAGR) and is projected to reach $2.45 billion by 2030 (research.md). Incumbent pricing ranges from one-time purchases (Paprika ~~$5) through subscriptions like Mealime Pro ($5.99/month), Plan to Eat ($49/year), and MyFitnessPal Premium+ (~~$20/month). Primary buyers are health-conscious households, families with mixed dietary needs, and users seeking to reduce food waste and grocery spending.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.