name: python-fastapi-ddd-tooling-skill description: Guides project tooling for a Python FastAPI + SQLAlchemy DDD/Onion Architecture codebase: uv-based environment setup, Makefile workflows, ruff formatting/linting, mypy typing, pytest, and CI (GitHub Actions), based on the dddpy reference. Use when bootstrapping a repo or tightening developer experience and quality gates. license: Apache-2.0 metadata: author: Takahiro Ikeuchi version: "1.0.0"

Tooling & DX for FastAPI DDD Projects (uv / ruff / mypy / CI)

This skill is about developer experience and quality automation for a DDD FastAPI project, following the conventions in dddpy.

Goals

• One-command local setup (make install)
• Fast feedback loop (make test, make format, make dev)
• Reproducible CI (GitHub Actions + Python matrix)

Recommended stack (dddpy-style)

• Python >=3.13
• uv for venv + dependency install + running tools
• ruff for formatting/linting
• mypy for type checking
• pytest for tests

Makefile workflow (core targets)

Keep a small Makefile that shells out to tools inside .venv/:

• venv: create .venv
• install: install editable package + dev deps
• test: run mypy + pytest
• format: run ruff formatter
• dev: run fastapi dev via uv

CI workflow

Run make install + make test on push/PR with a Python version matrix (e.g., 3.13, 3.14) and install uv in CI.

App bootstrap patterns

• Use FastAPI lifespan to create tables on startup and dispose the engine on shutdown.
• Keep SQLAlchemy engine/session setup in infrastructure/sqlite/database.py (or equivalent).
• Configure logging once at startup (logging.config.fileConfig(...)).

For concrete file templates (Makefile, workflow YAML, bootstrap snippets), read references/TOOLING.md.