Skip to main content
AI/MLwshobson

python-packaging

Create distributable Python packages with proper project structure, setup.py/pyproject.toml, and publishing to PyPI. Use when packaging Python libraries, creating CLI tools, or distributing Python code.

Stars
36,167
Source
wshobson/agents
Updated
2026-05-29
Slug
wshobson--agents--python-packaging
View on GitHubRaw SKILL.md

// install — copy + paste into any project

mkdir -p .claude/skills && curl -fsSL https://raw.githubusercontent.com/wshobson/agents/HEAD/plugins/python-development/skills/python-packaging/SKILL.md -o .claude/skills/python-packaging.md

Drops the SKILL.md into .claude/skills/python-packaging.md. Works with Claude Code, Cursor, and any agent that loads SKILL.md files from .claude/skills/.

Python Packaging

Comprehensive guide to creating, structuring, and distributing Python packages using modern packaging tools, pyproject.toml, and publishing to PyPI.

When to Use This Skill

  • Creating Python libraries for distribution
  • Building command-line tools with entry points
  • Publishing packages to PyPI or private repositories
  • Setting up Python project structure
  • Creating installable packages with dependencies
  • Building wheels and source distributions
  • Versioning and releasing Python packages
  • Creating namespace packages
  • Implementing package metadata and classifiers

Core Concepts

1. Package Structure

  • Source layout: src/package_name/ (recommended)
  • Flat layout: package_name/ (simpler but less flexible)
  • Package metadata: pyproject.toml, setup.py, or setup.cfg
  • Distribution formats: wheel (.whl) and source distribution (.tar.gz)

2. Modern Packaging Standards

  • PEP 517/518: Build system requirements
  • PEP 621: Metadata in pyproject.toml
  • PEP 660: Editable installs
  • pyproject.toml: Single source of configuration

3. Build Backends

  • setuptools: Traditional, widely used
  • hatchling: Modern, opinionated
  • flit: Lightweight, for pure Python
  • poetry: Dependency management + packaging

4. Distribution

  • PyPI: Python Package Index (public)
  • TestPyPI: Testing before production
  • Private repositories: JFrog, AWS CodeArtifact, etc.

Quick Start

Minimal Package Structure

my-package/
├── pyproject.toml
├── README.md
├── LICENSE
├── src/
│   └── my_package/
│       ├── __init__.py
│       └── module.py
└── tests/
    └── test_module.py

Minimal pyproject.toml

[build-system]
requires = ["setuptools>=61.0"]
build-backend = "setuptools.build_meta"

[project]
name = "my-package"
version = "0.1.0"
description = "A short description"
authors = [{name = "Your Name", email = "you@example.com"}]
readme = "README.md"
requires-python = ">=3.8"
dependencies = [
    "requests>=2.28.0",
]

[project.optional-dependencies]
dev = [
    "pytest>=7.0",
    "black>=22.0",
]

Package Structure Patterns

Pattern 1: Source Layout (Recommended)

my-package/
├── pyproject.toml
├── README.md
├── LICENSE
├── .gitignore
├── src/
│   └── my_package/
│       ├── __init__.py
│       ├── core.py
│       ├── utils.py
│       └── py.typed          # For type hints
├── tests/
│   ├── __init__.py
│   ├── test_core.py
│   └── test_utils.py
└── docs/
    └── index.md

Advantages:

  • Prevents accidentally importing from source
  • Cleaner test imports
  • Better isolation

pyproject.toml for source layout:

[tool.setuptools.packages.find]
where = ["src"]

Pattern 2: Flat Layout

my-package/
├── pyproject.toml
├── README.md
├── my_package/
│   ├── __init__.py
│   └── module.py
└── tests/
    └── test_module.py

Simpler but:

  • Can import package without installing
  • Less professional for libraries

Pattern 3: Multi-Package Project

project/
├── pyproject.toml
├── packages/
│   ├── package-a/
│   │   └── src/
│   │       └── package_a/
│   └── package-b/
│       └── src/
│           └── package_b/
└── tests/

Detailed patterns and worked examples

Detailed pattern documentation lives in references/details.md. Read that file when the navigation tier above is insufficient.