barkbarkgoose/environments

Environments

This skill detects what tools and runtimes a project uses, then routes to the appropriate spec file for working with those tools.

Detect Existing Environment

Before applying any setup, inspect the project to determine what's already in use:

| Signal | Inferred tool | |--------|--------------| | venv/, .python-version, requirements.txt | Python + uv virtualenv | | pyproject.toml + poetry.lock | Python + Poetry | | pyproject.toml + uv.lock | Python + uv (pyproject-based) | | package.json + package-lock.json | Node + npm | | package.json + yarn.lock | Node + Yarn | | package.json + pnpm-lock.yaml | Node + pnpm | | volta key in package.json | Node version managed by Volta | | .nvmrc or .node-version | Node version pinned | | docker-compose.yml or compose.yaml | Docker services defined | | Dockerfile present | App may be fully containerized |

Rule: Always detect before scaffolding. Treat existing artifacts as the source of truth.

Stack → Spec Mapping

These are the available spec files. Read only the ones relevant to the current stack:

| Environment | Spec file | When to read | |-------------|-----------|-------------| | Python (uv + virtualenv) | python-uv.md | Python backend — installing packages, running commands, scaffolding | | Node (npm) | node-npm.md | Node/Vue frontend — installing packages, running scripts | | Docker | docker.md | Service dependencies or full app containerization |

If the project uses a Python or Node tool not covered by a spec file (e.g. Poetry, Yarn, Volta), use the detected tool's own conventions and do not force the defaults from the spec files.

Key Rule

Each layer of the stack is always an independent environment — never mix Python package management with Node or vice versa. Each layer has its own .env file. See individual spec files for details.