MetaGPT: The Multi-Agent Framework

Assign different roles to GPTs to form a collaborative software entity for complex tasks.

1. MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.
2. Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs.
   1. Code = SOP(Team) is the core philosophy. We materialize SOP and apply it to teams composed of LLMs.

Software Company Multi-Role Schematic (Gradually Implementing)

Install

Pip installation

# Step 1: Ensure that Python 3.9+ is installed on your system. You can check this by using:
# You can use conda to initialize a new python env
#     conda create -n metagpt python=3.9
#     conda activate metagpt
python3 --version

# Step 2: Clone the repository to your local machine for latest version, and install it.
git clone https://github.com/geekan/MetaGPT.git
cd MetaGPT
pip3 install -e.     # or pip3 install metagpt  # for stable version

# Step 3: run the startup.py
# setup your OPENAI_API_KEY in key.yaml copy from config.yaml
python3 startup.py "Write a cli snake game"

# Step 4 [Optional]: If you want to save the artifacts like diagrams such as quadrant chart, system designs, sequence flow in the workspace, you can execute the step before Step 3. By default, the framework is compatible, and the entire process can be run completely without executing this step.
# If executing, ensure that NPM is installed on your system. Then install mermaid-js. (If you don't have npm in your computer, please go to the Node.js official website to install Node.js https://nodejs.org/ and then you will have npm tool in your computer.)
npm --version
sudo npm install -g @mermaid-js/mermaid-cli

detail installation please refer to cli_install

Docker installation

Note: In the Windows, you need to replace "/opt/metagpt" with a directory that Docker has permission to create, such as "D:\Users\x\metagpt"

# Step 1: Download metagpt official image and prepare config.yaml
docker pull metagpt/metagpt:latest
mkdir -p /opt/metagpt/{config,workspace}
docker run --rm metagpt/metagpt:latest cat /app/metagpt/config/config.yaml > /opt/metagpt/config/key.yaml
vim /opt/metagpt/config/key.yaml # Change the config

# Step 2: Run metagpt demo with container
docker run --rm \
    --privileged \
    -v /opt/metagpt/config/key.yaml:/app/metagpt/config/key.yaml \
    -v /opt/metagpt/workspace:/app/metagpt/workspace \
    metagpt/metagpt:latest \
    python startup.py "Write a cli snake game"

detail installation please refer to docker_install

QuickStart & Demo Video

• Try it on MetaGPT Huggingface Space
• Matthew Berman: How To Install MetaGPT - Build A Startup With One Prompt!!
• Official Demo Video

https://github.com/geekan/MetaGPT/assets/34952977/34345016-5d13-489d-b9f9-b82ace413419

Tutorial

• 🗒 Online Document
• 💻 Usage
• 🔎 What can MetaGPT do?
• 🛠 How to build your own agents?
  • MetaGPT Usage & Development Guide | Agent 101
  • MetaGPT Usage & Development Guide | MultiAgent 101
• 🧑‍💻 Contribution
  • Develop Roadmap
• 🔖 Use Cases
  • Debate
  • Researcher
  • Recepit Assistant
• ❓ FAQs

Support

Discard Join US

📢 Join Our Discord Channel!

Looking forward to seeing you there! 🎉

Contact Information

If you have any questions or feedback about this project, please feel free to contact us. We highly appreciate your suggestions!

• Email: alexanderwu@fuzhi.ai
• GitHub Issues: For more technical inquiries, you can also create a new issue in our GitHub repository.

We will respond to all questions within 2-3 business days.

Citation

For now, cite the arXiv paper:

@misc{hong2023metagpt,
      title={MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework}, 
      author={Sirui Hong and Mingchen Zhuge and Jonathan Chen and Xiawu Zheng and Yuheng Cheng and Ceyao Zhang and Jinlin Wang and Zili Wang and Steven Ka Shing Yau and Zijuan Lin and Liyang Zhou and Chenyu Ran and Lingfeng Xiao and Chenglin Wu and Jürgen Schmidhuber},
      year={2023},
      eprint={2308.00352},
      archivePrefix={arXiv},
      primaryClass={cs.AI}
}