README.md

How to Set Up a Python Project for Agentic Coding

This folder contains the sample code for the Real Python tutorial How to Set Up a Python Project for Agentic Coding.

It is the finished project from the tutorial: a small Python package wired up with the automated feedback an AI coding agent needs, namely Ruff for linting and formatting, mypy in strict mode for type checking, pytest with a coverage gate for behavior, and pre-commit to run those checks before every commit.

Setup

Install uv, then install the dependencies and the pre-commit hooks:

$ uv sync
$ uv run pre-commit install

Running the checks

Each tool can be run on its own:

$ uv run ruff check .
$ uv run mypy .
$ uv run pytest

Or run everything the way the agent is told to in AGENTS.md:

$ uv run pre-commit run --all-files

What each file is for

File	Purpose
pyproject.toml	Dependencies plus the Ruff, mypy, and pytest configuration
AGENTS.md	Workflow conventions an agent should follow
.pre-commit-config.yaml	Checks that run before each commit
tests/	Executable specification of the expected behavior
.env.example	Documents required environment variables without exposing secrets