MealMate - Your all-in-one meal planner, nutrition guide, and grocery assistant
This is a submission for the Agent.ai Challenge: Assembly of Agents (See Details)
What I Built
As part of this hackathon I have built MealMate which is like having a personal food assistant that combines three awesome agents —RecipePro, NutriBot, and PantryPal — into one easy agent. In order to acheive this I have used Spoonacular API which helps you find great recipes, check out their nutritional details, and make grocery lists with everything you need. It curates new meal ideas, track meal nutrition, or just figure out what to buy at the store, MealMate makes it all way easier. It’s built to save you time and keep things healthy, so you can enjoy cooking without the stress.
The bot has the following features.
- Finds the recipe
- Provide nutrition Analysis of the recipe
- Find wide variety of packed foods.
- Generate the audio support for MealMate article.
Now lets look at the demo.
Demo
In order to built MealMate I have created three small Agents.
- RecipePro Agent - https://agent.ai/agent/recipepro-agent
- NutriBot Agent - https://agent.ai/agent/nutribot
- PantryPal Agent - https://agent.ai/agent/pantrypal
My main submission is MealMate - https://agent.ai/agent/mealmate
Here is the demo of the working agent.
Agent.ai Experience
Agent.ai is very cool platform which provides great ease for the development of Agents. The documentation is really great along with video resources which helped me to ramp up quickly.
The action types provided covers lot of area and invoking agents and connecting them was easy and a smooth process.
In terms of the challenges faced was while playing around with actions I came across calling serverless functions and could not find a way to debug them from the Agent.ai platform. The hubspot blog does not seems to generate the article. When generating audio output action included the output is not generated in one go we need to click button twice.
Apart from the overall experience was great and I learnt a lot during the process.
Huge thanks to Dev Team and Agent.ai team for conducting this hackathon.
Source: View source
