ReAct Agent

A Python implementation of a ReAct (Reasoning and Acting) agent using LangChain and OpenAI. This agent can search Wikipedia and answer complex questions by reasoning through the search results step-by-step.

Features

• 🤖 ReAct (Reasoning and Acting) agent implementation
• 🔍 Wikipedia search and lookup capabilities
• 🧠 OpenAI GPT-3.5-turbo model integration
• 🔧 Environment variable configuration
• 📚 Professional project structure
• ✅ Fully tested and working

Prerequisites

• Python 3.7 or higher
• OpenAI API key

Setup

1. Clone or download this repository

2. Install dependencies

   pip install -r requirements.txt

3. Set up environment variables

   # Copy the example environment file
   cp .env.example .env
   
   # Edit .env and add your OpenAI API key
   nano .env

   Add your OpenAI API key to the .env file:

   OPENAI_API_KEY=your_actual_api_key_here
   

Usage

Run the ReAct agent:

python react-agent.py

The agent will process the example question about David Chanoff and the U.S. Navy admiral, demonstrating its reasoning and acting capabilities.

How it Works

1. Search Tool: The agent can search Wikipedia for relevant information
2. Lookup Tool: The agent can look up specific articles or sections
3. Reasoning: The agent uses the ReAct framework to reason about what information it needs
4. Acting: The agent takes actions (search/lookup) based on its reasoning
5. Answer: The agent provides a final answer based on the information gathered

Project Structure

ReAct Agent/
├── react-agent.py      # Main script
├── requirements.txt    # Python dependencies
├── .env.example       # Environment variables template
├── .env               # Your actual environment variables (not in git)
├── .gitignore         # Git ignore file
└── README.md          # This file

Dependencies

• openai>=1.6.1,<2.0.0: OpenAI API client
• wikipedia==1.4.0: Wikipedia API wrapper
• langchain==0.1.0: Framework for building LLM applications
• langchain-openai: OpenAI integration for LangChain
• langchain-community: Community tools and utilities
• langchainhub: LangChain Hub integration
• python-dotenv==1.0.0: Environment variable management

Example Output

🤖 ReAct Agent is thinking...
Question: Which Italian city was Michelangelo working in when he painted the ceiling of the Sistine Chapel, and what was the name of the Pope who commissioned this work?
--------------------------------------------------

> Entering new AgentExecutor chain...
I need to find out more about Michelangelo and the Sistine Chapel, specifically where he was working and who commissioned the work.
Action: Lookup
Action Input: Michelangelo
Observation: [Wikipedia search results about Michelangelo and his works]
I need to find more information about the Sistine Chapel and who commissioned it.
Action: Lookup
Action Input: Sistine Chapel
Observation: [Wikipedia search results about the Sistine Chapel, its location in Vatican City, and Pope Julius II's commission]
I now know that Michelangelo was working in Vatican City (Rome) when he painted the Sistine Chapel ceiling, and Pope Julius II commissioned this work.
Final Answer: Vatican City (Rome) and Pope Julius II

> Finished chain.
--------------------------------------------------
Answer: Vatican City (Rome) and Pope Julius II

Customization

To ask different questions, modify the question variable in react-agent.py:

question = "Your question here"

Example Questions for Testing

Here are some medium-difficulty questions that effectively demonstrate the ReAct agent's reasoning capabilities:

Art History: "Which Italian city was Michelangelo working in when he painted the ceiling of the Sistine Chapel, and what was the name of the Pope who commissioned this work?"

Martial Arts: "What is the name of the martial art style that Bruce Lee developed, and which traditional Chinese martial art did he study under Ip Man before creating his own system?"

Statistics: "Who developed the statistical method known as the t-test, and what was the name of the brewery where this statistician worked when he created this important statistical tool?"

These questions require multi-step reasoning, information synthesis, and demonstrate the agent's ability to connect different pieces of information from Wikipedia searches.

Notes

• Make sure to keep your .env file secure and never commit it to version control
• The agent uses the gpt-3.5-turbo model for cost-effectiveness
• Verbose mode is enabled to show the agent's reasoning process
• The agent uses a custom ReAct prompt template for Wikipedia research