CrewOS is an autonomous AI engineering runtime designed to orchestrate multi-agent collaboration, embedding memory, and multimodal capabilities. It provides a streamlined pipeline for planning, coding, reviewing, and refining AI projects.
- Modular agent orchestration
- Plan → Code → Review → Refine loop
- WebSocket-based real-time logs
- CLI and UI integration
- Embedding memory and feedback system
- Multi-crew collaboration for subprojects
- Python: 3.10+
- Dependencies: Listed in
setup.py
- Clone the repository:
git clone <repository-url>
cd crewos- Install the package:
pip install -e .- Set up environment variables:
Create a
.envfile in the root directory with the following:
OPENAI_API_KEY=your_openai_api_key
SERPER_API_KEY=your_serper_api_key
GH_TOKEN=your_github_token
-
Initialize a project with a template:
crewos init chatbot
-
Run the pipeline:
crewos run
-
Check the status of recent tasks:
crewos status
Run the following command to start the Streamlit-based UI:
streamlit run src/engineering/ui/app.pyThe UI provides real-time logs, memory traces, and task management.
.
│ .env
│ .gitignore
│ pyproject.toml
│ README.md
│
├── knowledge/
│ └── user_preference.txt
│
├── src/
│ └── engineering/
│ │ crew.py
│ │ main.py
│ │ __init__.py
│ │
│ ├── config/
│ │ └── agents.yaml
│ │ └── tasks.yaml
│ │
│ └── tools/
│ ├── custom_tool.py
│ └── __init__.py
│
└── tests/
└── test_pipeline.py
Run the unit tests using pytest:
pytest tests/Build and run the Docker container:
docker build -t crewos .
docker run -p 8501:8501 crewosContributions are welcome! Please fork the repository and submit a pull request.
This project is licensed under the MIT License. See the LICENSE file for details.