Scale computer-use 2.0 with open-source drivers, cross-OS sandboxes, and benchmarks for training, evaluation, and data generation.
Rust agentai-agentapplecomputer-usecomputer-use-agentcontainerizationcuadesktop-automationhacktoberfestlumemacosmanusoperatorswiftvirtualizationvirtualization-frameworkwindowswindows-sandbox
{ pkgs ? import <nixpkgs> {} }:
pkgs.mkShell {
buildInputs = with pkgs; [
# Python development (3.12 required by cua-bench)
python312
python312Packages.pip
python312Packages.virtualenv
# C/C++ libraries needed by numpy and other packages
stdenv.cc.cc.lib
# Docker for container operations
docker
docker-compose
# General tools
git
curl
jq
];
# Add library path for numpy and other native extensions
LD_LIBRARY_PATH = "${pkgs.stdenv.cc.cc.lib}/lib";
shellHook = ''
echo "CUA-Bench development environment loaded"
echo "Python version: $(python3 --version)"
# Create and activate venv if it doesn't exist
if [ ! -d ".venv" ]; then
echo "Creating virtual environment..."
python3 -m venv .venv
fi
source .venv/bin/activate
# Install package in editable mode if not already installed
if ! pip show cua-bench > /dev/null 2>&1; then
echo "Installing cua-bench in editable mode..."
pip install -e ".[dev]" 2>/dev/null || pip install -e .
fi
echo "Virtual environment activated"
'';
}