Devyan
Overview
Devyan is an AI-powered software development assistant that orchestrates a team of agents to solve programming tasks. It uses OpenAI's GPT-based agents to perform various roles such as architecture design, implementation, testing, and reviewing.
Features
- Architect Agent: Designs the architecture of the solution based on the user input.
- Programmer Agent: Implements the solution as per the architecture design.
- Tester Agent: Tests the implemented solution to ensure it meets the requirements and is free of bugs.
- Reviewer Agent: Reviews the architecture, implementation, and test results to provide a comprehensive analysis.
Requirements
- Python 3.7+
requests
librarylangchain
librarypython-decouple
librarycrewai
library- OpenAI API Key
Installation
-
Clone the repository:
git clone https://github.com/your-username/devain.git cd devain
-
Create a virtual environment:
python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate`
-
Install the dependencies:
pip install -r requirements.txt
-
Set up environment variables: Create a
.env
file in the root directory of the project and add your OpenAI API key:OPENAI_API_KEY=your_openai_api_key_here SERPER_API_KEY=your_serper_api_key_here
Usage
-
Run the main script:
python main.py
-
Follow the prompts:
## Welcome to Devain## ------------------------------- What problem do you want us to solve?
-
Enter the problem you want Devain to solve, and let the agents handle the rest.
Project Structure
devain/
│
│
├── tools/
│ ├── __init__.py
│ ├── file_write.py
│ └── directory_write.py
│
├── agents.py
│
├── tasks.py
│
├── main.py
│
├── requirements.txt
│
└── README.md
Contributing
- Fork the repository.
- Create your feature branch (
git checkout -b feature/your-feature
). - Commit your changes (
git commit -m 'Add some feature'
). - Push to the branch (`git push origin feature/your-feature``).
- Open a pull request.
Todo
- Create Custom tool to write file
- Create Custom tool to create directory
- Change prompts in tasks, make it more detialed and clear
- Use pytoml
- Add Agent Logs
- Stackoverflow tool
- Code Execution