Installation

Environment Setup

codemol uses Conda for dependency management, which manages compiled dependencies that pip alone cannot handle.

1. Clone the Repository

git clone https://github.com/rubencr14/codemol.git
cd codemol

2. Create the Conda Environment

conda env create -f environment.yml

This installs:

• PyQt5 — UI framework
• litellm — LLM integration for AI agents
• websockets — shared session relay

3. Activate

conda activate codemol

Common Issues

Import errors

If you get import errors after installation, ensure you activated the correct environment:

conda activate codemol
python -c "import codemol"

Qt platform plugin errors

On Linux without a display server:

export QT_QPA_PLATFORM=offscreen

Apple Silicon (M1/M2/M3)

Conda-forge packages support osx-arm64. If you encounter issues, try:

CONDA_SUBDIR=osx-arm64 conda env create -f environment.yml

Optional: AI Agent Support

To use the AI agent features, set up an API key:

OpenAIAnthropic

export OPENAI_API_KEY="sk-..."
export AI_PROVIDER="openai"

export ANTHROPIC_API_KEY="sk-ant-..."
export AI_PROVIDER="anthropic"

Verify Installation

conda activate codemol
python main.py

You should see the two-panel window: 3D viewer on top, console on the bottom.