AI ReAct Agent

A Streamlit-based web application that implements a ReAct (Reasoning and Acting) agent using LangChain and OpenAI. The agent can search information, fetch stock market data, and get weather updates.

Features

• Interactive chat interface using Streamlit
• ReAct agent implementation with LangChain
• Integration with multiple APIs:
  • Tavily Search
  • Alpha Vantage (stock market data)
  • WeatherAPI (weather information)
  • OpenAI GPT-3.5 Turbo

Prerequisites

• Python 3.8+
• OpenAI API key
• LangChain API key
• Tavily API key
• Alpha Vantage API key
• WeatherAPI key

Installation

1. Clone the repository:

git clone https://github.com/ferrerallan/ai-react-agent.git
cd ai-react-agent

2. Install dependencies:

pip install -r requirements.txt

3. Configure environment variables:

• Copy .env.example to .env
• Fill in your API keys:

OPENAI_API_KEY=your_key
LANGCHAIN_API_KEY=your_key
LANGCHAIN_TRACING_V2=true
LANGCHAIN_PROJECT=your_project
TAVILY_API_KEY=your_key
ALPHA_VANTAGE_KEY=your_key
WEATHERAPI_KEY=your_key

Running the Application

Start the Streamlit app:

streamlit run run.py

Project Structure

• run.py: Main application entry point with Streamlit interface
• nodes.py: Defines agent reasoning engine and tool execution
• react.py: Implements ReAct agent and tools (stock data, weather)
• state.py: Defines agent state management

Available Tools

1. Stock Market Data

   • Fetches intraday stock data with 5-minute intervals
   • Provides open, high, low, close prices and volume

2. Weather Information

   • Gets current weather conditions for any city
   • Shows temperature, conditions, and wind information

3. Web Search

   • Performs web searches using Tavily API
   • Returns relevant search results