Compare Python Project Scaffolding: Cookiecutter, PyScaffold, PyBuilder, Poetry

Python has long lacked a de facto standard project management and build tool, resulting in diverse structures.

Unlike Java, which standardized on Maven (and alternatives like Gradle, SBT, etc.), Python's ecosystem offers package managers (pip, pipenv, conda) but no enforced layout.

Typical Python projects still rely on Makefiles, setup.py, build.py, or template tools.

Cookiecutter – a classic Python project layout

$ pip install cookiecutter
$ cookiecutter gh:audreyr/cookiecutter-pypackage
# answer prompts, e.g. project_name [Python Boilerplate]: sample

Running cookiecutter gh:audreyr/cookiecutter-pypackage generates a project with the following structure:

sample
├── AUTHORS.rst
├── CONTRIBUTING.rst
├── HISTORY.rst
├── LICENSE
├── MANIFEST.in
├── Makefile
├── README.rst
├── docs
│   ├── Makefile
│   ├── authors.rst
│   ├── conf.py
│   ├── contributing.rst
│   ├── history.rst
│   ├── index.rst
│   ├── installation.rst
│   ├── make.bat
│   ├── readme.rst
│   └── usage.rst
├── requirements_dev.txt
├── sample
│   ├── __init__.py
│   ├── cli.py
│   └── sample.py
├── setup.cfg
├── setup.py
├── tests
│   ├── __init__.py
│   └── test_sample.py
└── tox.ini

The layout includes a sample source package, a tests directory, docs, and build files such as setup.py and Makefile. Build can be driven by make commands (clean, lint, test, coverage, docs, dist, install).

PyScaffold – creating a project

$ pip install pyscaffold
$ putup sample

PyScaffold creates a similar layout but places source code under a src directory.

sample
├── AUTHORS.rst
├── CHANGELOG.rst
├── CONTRIBUTING.rst
├── LICENSE.txt
├── README.rst
├── docs
│   ├── Makefile
│   ├── _static
│   ├── authors.rst
│   ├── changelog.rst
│   ├── conf.py
│   ├── contributing.rst
│   ├── index.rst
│   ├── license.rst
│   ├── readme.rst
│   └── requirements.txt
├── pyproject.toml
├── setup.cfg
├── setup.py
├── src
│   └── sample
│       ├── __init__.py
│       └── skeleton.py
├── tests
│   ├── conftest.py
│   └── test_skeleton.py
└── tox.ini

Building and testing is performed with tox, which creates isolated virtual environments.

$ tox -av
# shows default, build, clean, docs, doctests, linkcheck, publish, etc.

Specific tasks can be run with tox -e build, tox -e docs, and so on.

PyBuilder

$ pip install pybuilder
$ mkdir sample && cd sample
$ pyb --start-project
# answer prompts, project files are created

PyBuilder generates a Maven‑like structure and defines tasks in build.py. Common tasks include analyze, compile_sources, test, package, and publish. Tasks can be listed with pyb -t and executed with pyb.

Poetry

$ pip install poetry
$ poetry new sample

Poetry creates a minimal layout and stores all configuration in pyproject.toml, which also manages dependencies (e.g., poetry add boto3) and builds wheels.

$ poetry build
$ poetry shell
$ poetry run pytest

Poetry does not generate documentation or coverage reports automatically; those must be invoked via poetry run with tools such as Sphinx or coverage.

Overall, the complexity of the generated project structure decreases from Cookiecutter → PyScaffold → PyBuilder → Poetry, while the learning curve follows the same order.