AI Meal Planner

Generates weekly meal plans by filtering recipes against user-specified allergies, intolerances, and dietary preferences (vegetarian, vegan, keto, etc.) using constraint-satisfaction logic. The system maintains a curated recipe database tagged with ingredient metadata and nutritional profiles, then applies multi-constraint filtering to ensure no conflicting ingredients appear in generated plans. This approach differs from generic meal planners by enforcing hard constraints rather than soft recommendations, preventing accidental allergen exposure.

Extracts and aggregates nutritional data (calories, macros, micronutrients) from individual recipes and presents weekly summaries alongside meal plans. The system likely uses a pre-computed nutrition database (USDA or proprietary) linked to recipe ingredients, calculating totals by summing ingredient nutrition facts. This differs from recipe-only tools by surfacing nutrition as a primary output, not a secondary lookup, enabling users to validate plans against dietary goals.

Structures generated meals into a 7-day calendar view with 3 meals per day (breakfast, lunch, dinner) and optional snacks, presenting recipes with links to full instructions and ingredient lists. The system uses a template-based layout engine that maps recipes to day/meal slots, likely with basic conflict detection to avoid recipe repetition within a week. This differs from recipe search tools by providing a ready-to-execute weekly structure rather than requiring manual assembly.

Accepts user preferences (cuisine type, cooking time, ingredient preferences) as input filters and biases recipe selection toward matching preferences during plan generation. The system likely uses a preference-weighting algorithm that scores recipes based on user inputs (e.g., 'quick meals' → prioritize recipes under 30 minutes, 'Mediterranean' → weight Mediterranean recipes higher) before constraint filtering. This differs from static meal plans by tailoring recommendations to individual taste and lifestyle constraints.

Provides a searchable interface to Casa de Sante's recipe database with filters for ingredients, dietary tags, prep time, and nutritional criteria. The system likely uses full-text search (Elasticsearch or similar) combined with faceted filtering to enable users to browse recipes independently of meal plan generation. This differs from meal-plan-only tools by offering recipe discovery as a standalone feature, allowing users to explore options before committing to a full week.

Aggregates ingredients from all recipes in a generated meal plan and produces a consolidated grocery list, optionally organized by store section (produce, dairy, pantry) or by recipe. The system deduplicates ingredients across recipes (e.g., if 'olive oil' appears in 3 recipes, it is listed once with combined quantity) and likely exports to text, PDF, or CSV formats. This differs from manual list-making by automating ingredient aggregation and reducing shopping friction.

Maintains a user profile with declared allergies, intolerances, and sensitivities (e.g., peanut allergy, lactose intolerance, FODMAP sensitivity) and applies these constraints to all meal plan generation and recipe recommendations. The system stores allergen data in a user profile (likely relational database) and cross-references against recipe ingredient metadata during filtering. This differs from single-use allergen filters by persisting preferences across sessions and ensuring consistent safety enforcement.

Curates and tags recipes specifically for FODMAP compliance and digestive health, using Casa de Sante's proprietary nutritional science database to identify low-FODMAP ingredients and preparation methods. The system likely maintains a separate 'gut-health' recipe subset with additional metadata (FODMAP level, trigger ingredients, digestive impact) beyond standard recipe data. This differs from generic meal planners by applying medical nutrition science to recipe selection, not just allergen avoidance.