ManualTrigger Automate

For ManualTrigger Automate, effortlessly fetch and store HelloFresh's weekly menu, build a recipe recommendation engine using Qdrant, and interact with an AI agent for personalized meal suggestions. This workflow simplifies meal planning by integrating real-time data and user preferences, ensuring you receive tailored recipe recommendations based on your tastes.

7/8/2025
33 nodes
Complex
manualcomplexlangchainexecuteworkflowtriggerwaitsticky noteadvancedapiintegration
Categories:
Complex WorkflowManual TriggeredBusiness Process Automation
Integrations:
LangChainExecuteWorkflowTriggerWaitSticky Note

Target Audience

This workflow is ideal for:
- Home Chefs: Individuals looking for meal inspiration and recipe recommendations tailored to their preferences.
- Food Enthusiasts: Those who enjoy exploring new cuisines and categories of dishes.
- Healthy Eaters: Users who want to track nutrition and ingredient details for healthier meal choices.
- Developers and Data Scientists: Professionals interested in integrating AI and machine learning into culinary applications, utilizing APIs for recipe recommendations.
- Businesses: Companies in the food industry looking to automate recipe suggestions and enhance customer engagement through personalized meal planning.

Problem Solved

This workflow addresses the challenge of finding suitable recipes based on current preferences and weekly menus. It automates the process of scraping HelloFresh's weekly menu, extracting detailed recipe information, and generating personalized recommendations using AI. This eliminates the need for users to manually search for recipes, streamlining meal planning and enhancing the overall cooking experience.

Workflow Steps

  • Manual Trigger: The workflow begins with a manual trigger, allowing users to initiate the process by clicking a button.
    2. Fetch Weekly Menu: It retrieves the current week's HelloFresh menu using an HTTP request.
    3. Extract Data: The workflow extracts relevant data from the HTML content, including available courses and recipe metadata.
    4. Get Recipe Details: For each course, it fetches detailed recipe information, including description, ingredients, and instructions.
    5. Prepare Documents: The workflow prepares structured documents for each recipe, including nutritional information and cooking steps.
    6. Vectorization: Recipes are converted into vector embeddings using Mistral Cloud, preparing them for a recommendation engine.
    7. Store in Qdrant: The vectorized recipes are stored in a Qdrant vector store, enabling efficient retrieval based on user preferences.
    8. Integrate AI Agent: An AI agent is set up to interact with users, providing personalized recipe recommendations based on their likes and dislikes.
    9. Execute Workflow: The workflow can be executed to fetch recommendations based on user input, utilizing the stored data and embeddings.

    • Customization Guide

    Users can customize this workflow by:
    - Modifying API Endpoints: Change the URL in the HTTP request node to fetch recipes from different sources or customize the HelloFresh endpoint.
    - Adjusting Data Extraction: Update the CSS selectors in the HTML extraction nodes to target different elements based on website structure changes.
    - Personalizing Recipe Filters: Customize the AI agent's system message to reflect specific dietary preferences or restrictions.
    - Changing Recommendation Criteria: Adjust the parameters used in the Qdrant Recommend API to refine how recommendations are generated based on user feedback.
    - Enhancing Database Storage: Modify the database schema or add additional fields in the SQLite storage to capture more information about recipes.