Autogen experiments and playground
autogen-playground is a repository for experimenting with the Microsoft autogen framework. The main technical feature of this environment is the usage of "agents" with specific functionalities. These agents collaboratively work to perform specific tasks.
To get a local copy running, follow these steps:
- Clone the repository
git clone https://github.com/msaelices/autogen-playground.git
- Install the required packages
pip install -r requirements.txt
- Make sure to setup your environment variables as per the configuration of the project
cp src/.env-default src/.env # and fill the environment variables
cp src/OAI_CONFIG_LIST.sample src/OAI_CONFIG_LIST # and set the OpenAI key
This program is designed to be run from the command line with a single task string as input. You initiate the main.py
and input the task you want to execute.
$ cd src/
$ python main.py
Task to implement: Your task here
Each agent in the program has its own role:
user-proxy
: Initiates the chat with a task messageresearcher
: Fetches data from the internet using the google search enginecode-expert
: Provides coding expertise for the task at handproduct-manager
: Manages the interaction between the other agents, help break down the initial task into a well scoped requirement
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.