private-gpt/open-devin

An Open Source Multi-Agent LLM Software Dev inspired by Devin

Open-Devin AI Software Engineer

Overview

Open-Devin is an AI-powered software engineering system designed to automate and streamline complex development tasks. Leveraging advanced AI agents, this system facilitates web data scraping, database management, code generation, and interaction with GitHub repositories.

Features

• Chain-of-Thought Planner (CoT): Employs a strategic AI agent for constructing and executing action plans, guiding complex problem-solving tasks.
• Web Scraper: Automatically gathers data from the web, focusing on trusted and relevant sources.
• SQLite Database: Provides a robust storage system for data persistence and retrieval, acting as a dynamic knowledge base.
• Code Generation: Executes Python and shell code within an interpreter, supporting a vast range of software engineering tasks.
• GitHub Interaction: Manages issues, pull requests, and content within GitHub repositories.
• Isolated Execution Environments: Offers secure code execution within containerized environments like Docker.
• Synthesis Microservice: Integrates outputs from various agents, providing synthesized and actionable responses.

Getting Started

To begin using Open-Devin, you'll need to:

1. Clone the repository:

   git clone https://github.com/yourusername/open-devin.git

2. Install the required dependencies:

   pip install -r requirements.txt

3. Run the system:

   python main.py

4. Interact with the system through the command-line interface. Set up the necessary environment variables for API keys and database connections.

Contributing

Contributions to Open-Devin are welcome! If you would like to contribute:

1. Fork the repository.
2. Create a new branch for your feature or bug fix.
3. Make your changes and commit them.
4. Push your changes to your fork.
5. Open a pull request to the main repository.

License

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

Acknowledgements

• LangChain for the LLM capabilities.
• PyGithub for GitHub API interactions.
• All the contributors who have invested time in providing feedback and suggestions.