MCTF Software Setup and Installation

MCTF uses the Pyquaticus game simulation environment. Pyquaticus is an open-source, Python-based library developed at MIT Lincoln Laboratory that enables multi-agent training via reinforcement learning (RL) using a PettingZoo/Gymnasium environment. It supports the RLlib deep RL library and can integrate with other libraries such as Stable-Baselines or custom RL implementations. Pyquaticus also supports heuristic-based agents and manual keyboard control of agents.

Steps to Download and Install Pyquaticus

  1. Download Pyquaticus from the GitHub repository.
  2. git checkout mctf2026 (NOTE mctf2026 is the branch that will be used for the 2026 competition not main)
  3. Follow the instructions in the README.md file.
  4. Set up a Python virtual environment using one of the methods below.
  5. Train your first RL agents using the Getting Started guide.

Setting up a Python Virtual Environment

Option 1: Using Miniconda (Recommended)

  1. Install Miniconda from here.
  2. Navigate into the cloned Pyquaticus repository.
  3. Run:
    ./setup-conda-env.sh light
    ./setup-conda-env.sh full (recommended, includes RLlib)

Option 2: Using Python Virtual Environment

  1. Install Python 3.10.
  2. Create a virtual environment:
    python3.10 -m venv <envname>
  3. Install Pyquaticus:
    pip install -e .[torch,ray] (full)
    or pip install -e . (light)

Setting Up on Windows

  1. Install Miniconda from here.
  2. Clone the Pyquaticus GitHub repository.
  3. Navigate into the repository.
  4. git checkout mctf2026 (NOTE mctf2026 is the branch that will be used for the 2026 competition not main)
  5. Remove line 56 (pymoos==2022.1) from pyproject.toml.
  6. Run:
    ./setup-conda-env.sh light (WSL)
    or
    ./setup-conda-env.sh full (includes RLlib)
  7. Or install manually:
    pip install -e .[torch,ray]
    or
    pip install -e .