MultiAgentToolbox

Multi-Agent Toolbox 🤖

Transform your queries into actionable insights with AI-powered agents! This Streamlit application leverages LangChain and Groq to create specialized agents that handle weather forecasts and mathematical calculations.

Features ✨

• Weather Forecasting Agent: Get real-time weather information for any city worldwide
• Mathematical Calculation Agent: Perform complex calculations and solve equations
• Modern Streamlit UI: Beautiful gradient design with responsive cards and animations
• Multi-Server Architecture: Independent agents for different domains
• Error-Resistant Workflow: Comprehensive error handling and user feedback
• Responsive Design: Works seamlessly on desktop and mobile devices

Requirements 📋

Python 3.9+

Streamlit

LangChain

LangGraph

FastMCP

Groq API Key

OpenWeather API Key

Configuration

The system uses specialized agents with distinct configurations:

Math Agent: Runs locally through Python, executing math operations with step-by-step explanations

Weather Agent: Connects to OpenWeather API via HTTP to fetch real-time forecasts

Agent Controller: Manages communication between agents using LangGraph

API Integration: Requires Groq API (for Llama3 AI model) and OpenWeather API keys stored in .env file

How It Works

User Query Processing:

Users submit natural language queries through the Streamlit web interface

The system analyzes whether the query requires math calculation, weather data, or both

Agent Execution:

Math queries route to the local math server that performs calculations

Weather queries call the weather server which fetches live data from OpenWeather

Response Generation:

Math operations return detailed step-by-step solutions

Weather forecasts provide temperature and conditions

Combined queries show conversion processes between units

Results display in a formatted UI with visual feedback

Error Handling:

System catches API failures, math errors (like division by zero), and invalid inputs

User-friendly error messages explain issues clearly