If you open any well-maintained Python project on GitHub, you’ll notice that .py files are almost outnumbered by everything else: configuration files, dotfiles, a Dockerfile (and sometimes a Makefile), which you may or may not ever use.

You know the source code matters, but what is all of this other stuff, and do you actually need it, and if you do, what do you need to include especially with the advancement of AI?

Here, I’m going to provide a few GitHub templates as well as a breakdown of some of the major files that you’d find in a repository.

GitHub has a “Use this template” button that lets you clone a repo’s structure into a new project without forking it. There are some well established Python templates worth knowing about:

These are great “out of the box” tools, as they’ll save you time. But they all have a single blind spot: none of them account for AI-assisted development.

If you’re using Claude Code (or planning to), I created a repository with a pre-built CLAUDE.md file that uses modern practices (such as uv for package management with pre-loaded packages Claude uses) called python-project-template-with-claude.

» Note: it’s an open source project in the early stages - I’d love for you to contribute!

The CLAUDE.md file contains guardrails of do’s and dont’s as well as best practices for the model to follow.

The template feels lean, which is allowing for versatility - it’s impossible to anticipate project types. It’s giving you a clean starting point where both you and your AI tooling are on the same page from commit zero.

Let’s break down what belongs in a project repository; you’ll see the majority of these components in the python-project-template-with-claude repository (at the time of writing).

This is a file defining your project’s name, version, dependencies, and build system. Older projects may use setup.py and setup.cfg, but the new standard for 2026 is using pyproject.toml.

Virtual environments are a core to all Python projects, as they manage dependencies you may use. Modern practices use uv, but many projects still use pip or Poetry.

uv uses the pyproject.toml file to manage it’s dependencies, but you may run into a requirements.txt file if the project uses pip, or an environment.yaml file if the project uses Conda.

» To be clear, uv and pip is the tool to manage your dependencies - the dependencies are written in either pyproject.toml or requirements.txt.

If you’re using git, you’re going to want to ignore files - this can be accomplished by adding a .gitignore file to the project repository.

This is a file that contains secrets like API keys, login information, etc. It’s very important to note that this does NOT get checked into source control (i.e. git) and is added to the .gitignore file.

Sometimes, you need more than a README.md file. You may see a docs directory, which will house all the documentation for the project. Oftentimes, you’ll see projects use tools like MkDocs or Sphinx to manage the documentation.

Depending upon how the project is structured, you may see a tests folder with a test configuration file, such as pytest.ini. Modern practices add a [tool.pytest.ini_options] in the pyproject.toml file.

When you breach into containerization, you may see a Dockerfile (or a docker-compose.yaml file if there’s multiple services). I wrote about deploying your Python project using Docker here with a minimal file.

A lot of projects also include a .github/workflows directory for GitHub actions (or .gitlab-ci.yaml for GitLab projects). These files include testing, linting, and deployment on every push or pull request to automate actions.

This field is moving very quickly, but here’s a listing of a few files that you may see in a repository:

Given how quick the landscape is evolving, I highly encourage you to use the python-project-template-with-claude repository and make contributions to make the project more solid!

Happy coding!

Want even more Python-related content that’s useful? Here’s 3 reasons why you should subscribe the Python Snacks newsletter:

You can unsubscribe at any time.

Do you have a wealth of knowledge and insights to share with the world? Starting your own newsletter or blog is an excellent way to establish yourself as an authority in your field, connect with a like-minded community, and open up new opportunities.

If TikTok, Twitter, Facebook, or other social media platforms were to get banned, you’d lose all your followers. This is why you should start a newsletter: you own your audience.

» Use this link to get 20% off for 3 months your newsletter! «

This article may contain affiliate links. Affiliate links come at no cost to you and support the costs of this blog. Should you purchase a product/service from an affiliate link, it will come at no additional cost to you.