Setting up the Python environment for the Copilot Agent

[EveCharbie]

Select Topic Area

Question

Copilot Feature Area

Copilot Agent Mode

Body

I have followed the recommendations here (https://docs.github.com/en/copilot/customizing-copilot/customizing-the-development-environment-for-copilot-coding-agent#preinstalling-tools-or-dependencies-in-copilots-environment).
So, I created a copilot-setup-steps.yml that is successfully running on my repo's CI. However, the Copilot Agent still struggles creating an appropriate environment (pyomeca/biobuddy#29). Did someone succeed in providing a Python environment to Copilot ?

[IT22194930]

✅ Things to Check in Your copilot-setup-steps.yml:

1. Use Conda or venv properly
   Make sure you're installing Python using conda, pyenv, or actions/setup-python, and activating the environment correctly.

2. Install dependencies globally in the environment
   After activating the environment, install your dependencies:

   pip install -r requirements.txt
   

3. Ensure Copilot Agent uses the correct Python interpreter
   Sometimes Copilot may default to the system Python. Explicitly set the PYTHONPATH or confirm the path:

   export PYTHONPATH=$(pwd)
   which python
   

4. Add environment verification step
   Confirm everything is installed by printing version and dependencies:

   python --version
   pip list

[EveCharbie]

The fix was to use 'activate-environment' in the conda setup :)

- name: Setup conda
    uses: conda-incubator/setup-miniconda@v3
    with:
      miniforge-version: latest
      activate-environment: name_of_the_env
      python-version: 3.11.11