Quickstart

This guide will help you get started with AgentOpera’s basic functionality.

📋 Prerequisites

Before starting, make sure you have:

1. Installed AgentOpera (see installation guide)

2. Set up your environment variables for LLM API access

🔑 Setting Up Environment Variables

AgentOpera uses environment variables for API keys and model configurations.

Configure them as follows:

# OpenAI API Configuration
export MODEL_BASE_URL=https://api.openai.com/v1
export MODEL_API_KEY=your_openai_api_key
export MODEL_ID=gpt-4o  # or any other model ID you prefer

👋 Hello World Example

Start with a simple “Hello World” example:

import asyncio
import os
from agentopera.chatflow.agents import AssistantAgent
from agentopera.models.openai import OpenAIChatCompletionClient

# Create a model client using the OpenAI API key and model ID
model_client = OpenAIChatCompletionClient(
    base_url=os.getenv("MODEL_BASE_URL"),
    api_key=os.getenv("MODEL_API_KEY"),
    model=os.getenv("MODEL_ID")
)

async def main() -> None:
    # Create an assistant agent using the model client
    agent = AssistantAgent("assistant", model_client=model_client)
    # Run the agent with the task "Say 'Hello World!'"
    print(await agent.run(task="Say 'Hello World!'"))

asyncio.run(main())

Save this code to a file (e.g., hello_world.py) and run it:

python hello_world.py

🚀 Next Steps

For more advanced usage, including:

• Creating complex agent networks

• Implementing custom agents

• Deploying agents as services

Refer to our complete documentation.

👏 Acknowledgements

We thank AutoGen by Microsoft for their excellent work, which inspired our Python API implementation. 🔗 Original Project: AutoGen