- 2024.3: 🎉 OS-Copilot is accepted at the LLM Agents Workshop@ICLR 2024.
OS-Copilot is an open-source library to build generalist agents capable of automatically interfacing with comprehensive elements in an operating system (OS), including the web, code terminals, files, multimedia, and various third-party applications.
-
Clone the GitHub Repository:
git clone https://github.com/OS-Copilot/OS-Copilot.git
-
Set Up Python Environment and Install Dependencies:
conda create -n oscopilot_env python=3.10 -y conda activate oscopilot_env cd OS-Copilot pip install -e .
-
Set OpenAI API Key: Configure your OpenAI API key in .env.
cp .env_template .env
-
Now you are ready to have fun:
python quick_start.py
* FRIDAY currently only supports single-round conversation.
Level | Tutorial | Description |
---|---|---|
Beginner | Installation | Explore three methods to install FRIDAY. |
Beginner | Getting Started | The simplest demonstration of FRIDAY with a quick_start.py script. |
Intermediate | Adding Your Tools | Adding and removing tools to the FRIDAY. |
Intermediate | Deploying API Services | Demonstrate the deployment of API services for FRIDAY. |
Intermediate | Example: Automating Excel Tasks | Automating Excel control using FRIDAY. |
Intermediate | Enhancing FRIDAY with Self-Learning for Excel Task Automation | Improved Excel control with self-directed learning. |
Advanced | Designing New API Tools | Guides on deploying custom API tools for FRIDAY to extend its functionalities. |
Join our community to connect with other enthusiasts, researchers and developers:
- Discord: Join our Discord server for real-time discussions and support.
- Twitter: Follow our Twitter to get latest new or tag us to share your demos!
Visit the roadmap to preview what the community is working on and become a contributor!
OS-Copilot is provided "as is" without warranty of any kind. Users assume full responsibility for any risks associated with its use, including potential data loss or changes to system settings. The developers of OS-Copilot are not liable for any damages or losses resulting from its use. Users must ensure their actions comply with applicable laws and regulations.
@misc{wu2024oscopilot,
title={OS-Copilot: Towards Generalist Computer Agents with Self-Improvement},
author={Zhiyong Wu and Chengcheng Han and Zichen Ding and Zhenmin Weng and Zhoumianze Liu and Shunyu Yao and Tao Yu and Lingpeng Kong},
year={2024},
eprint={2402.07456},
archivePrefix={arXiv},
primaryClass={cs.AI}
}
If you have any inquiries, suggestions, or wish to contact us for any reason, we warmly invite you to email us at wuzhiyong@pjlab.org.cn.