README.md

Hello World Agent

A simple demonstration agent using the ReACT methodology for analyzing and executing tasks.

Quick Install

pip install hello_agent

Prerequisites

• Python 3.8 or higher
• OpenRouter API key (for LLM access)

Installation

1. Install via pip:

   pip install hello_agent

   Or install from source:

   git clone https://github.com/ruvnet/hello_world_agent.git
   cd hello_world_agent
   pip install -e .

2. Add your OpenRouter API key to the .env file:

   OPENROUTER_API_KEY=your_api_key_here

Usage

1. Start the agent using the command-line tool:

   agent --prompt "What is quantum computing?" --task research

   Or run from source:

   ./start.sh --prompt "What is quantum computing?" --task research

Command Line Arguments

• --prompt: Specify the input prompt (default: "Tell me about yourself")
• --task: Specify the task type: research, execute, analyze, or both (default: both)
• --hitl: Enable human-in-the-loop mode (optional)

Example:

agent --prompt "What is quantum computing?" --task research --hitl

Features

• ReACT Methodology Implementation
• Research Analysis
• Task Execution
• Performance Analysis
• Progress Tracking
• Streaming Responses

Project Structure

agent/
├── config/              # Configuration files
│   ├── agents.yaml     # Agent definitions
│   ├── tasks.yaml      # Task definitions
│   └── analysis.yaml   # Analysis rules
├── tools/              # Custom tools
├── docs/               # Documentation
└── examples/           # Example implementations

Documentation

For detailed documentation and user guides, refer to:

• User Guide
• Templates Guide
• Tools Guide
• Configuration Guide
• Advanced Implementations Guide
• Memory and Storage Guide
• Human-in-the-Loop Guide

Examples

Explore the Examples directory for sample usage scenarios and human-in-the-loop implementations.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Built with CrewAI
• Powered by OpenRouter