Python Environments

Owns one question for the whole workspace: which Python are we using? It finds the environments on your machine, creates new ones, installs packages into them, and publishes the active one so every Python-aware extension runs against the same interpreter instead of guessing python3.

What you can do

• Find the environments you already have: workspace venvs (.venv, venv, .env, env), conda and mamba environments, pyenv versions, poetry and pipenv environments, and system interpreters
• Create a venv (with uv when available, python -m venv otherwise) or a conda environment, with packages installed as part of the setup
• Install and uninstall packages in whichever environment you pick, always through that interpreter rather than a bare pip from PATH
• Switch the active environment and have Local Jupyter, the debugger, Requirements, and Experiments all follow it
• Open a terminal that is already activated in an environment, with no activate script to remember
• Get a real way forward when a system Python refuses installs, instead of a externally-managed-environment wall

Where to find it

• The Python Environments sidebar lists your environments with the active one marked, over the packages installed in it. Right-click a row for Use, Open Activated Terminal, Install, Reveal in Finder, or Delete. The + in each section header creates an environment or installs a package.
• The status bar shows the active environment. Click it to switch.
• A filterable Select Python Environment picker is available from anywhere.

The active environment is remembered in .punica/python.yaml, a plain file you can edit by hand.

Settings

• env-manager.pythonPath — an explicit interpreter path that always wins over discovery. Use it for an environment this extension cannot find.

Requirements

The desktop app only: discovery and creation run local processes.

Getting started

Open the Python Environments sidebar and pick an environment, or press + to create a .venv for this workspace.