Skip to content

1. Installation

Prerequisites

Before installing CMA, ensure you have the following prerequisites:

  • Python 3.12 or later

    CMA uses multiprocessing and asyncio, which are fully supported in Python 3.12 and later versions.

  • Docker

    CMA uses Docker to run modules (specifically, MQTT for messaging and ChromaDB for vector storage). Ensure you have Docker installed and running on your machine. Please refer to the Docker installation guide for instructions.

Setup

It is recommended to use a virtual environment:

$ python -m venv .venv
$ source .venv/bin/activate

Clone this repository and install the required packages:

$ git clone https://github.com/***.git
$ cd concurrent-modular-agent-main
$ pip install -e ".[dev]"

Testing Your Installation

1. Start Backend Services

$ coma backend start

Note: You need to run Docker before starting the backend services.

Tip for macOS users: You need to run the Docker Desktop application, not CLI commands.

You can use the start, stop, and restart commands to manage the backend services. For example, to stop the backend services:

$ coma backend stop

2. Set Your OpenAI API Key

$ export OPENAI_API_KEY="your_api_key"

3. Run Tests

$ pytest

If all tests pass, you should see output similar to: 

....

If tests fail, you will see output indicating which tests failed or encountered errors: 

..!!!

Next Steps

Now you're ready to start building your agent systems with CMA! Continue to the next section to learn how to create your first project.