Python Best Practices

When to Use This Skill

This skill should be triggered when:

•Writing or reviewing Python code
•Setting up Python projects or dependencies
•Configuring linting, formatting, or type checking
•Discussing Python patterns and conventions
•Working with FastAPI, Click/Typer, or any Python framework

Core Capabilities

•Package Management: uv for fast, reliable dependency management
•Code Quality: ruff for linting and formatting (replaces black, isort, flake8)
•Type Checking: ty for type validation
•Project Structure: Shared core pattern for API + CLI projects

Package Management with uv

Why uv

•10-100x faster than pip
•Replaces pip, pip-tools, virtualenv, and pyenv
•Lockfile support for reproducible builds
•Built-in Python version management

Common Commands

bash

# Create new project
uv init my-project
cd my-project

# Add dependencies
uv add fastapi uvicorn
uv add --dev pytest ruff

# Sync dependencies (install from lockfile)
uv sync

# Run commands in venv
uv run python script.py
uv run pytest

# Pin Python version
uv python pin 3.12

pyproject.toml Structure

toml

[project]
name = "my-project"
version = "0.1.0"
requires-python = ">=3.12"
dependencies = [
    "fastapi>=0.115.0",
    "uvicorn>=0.32.0",
]

[project.optional-dependencies]
dev = [
    "pytest>=8.0.0",
    "ruff>=0.8.0",
]

[tool.uv]
dev-dependencies = [
    "pytest>=8.0.0",
    "ruff>=0.8.0",
]

Code Quality with ruff

Configuration

toml

[tool.ruff]
line-length = 88
target-version = "py312"

[tool.ruff.lint]
select = [
    "E",      # pycodestyle errors
    "W",      # pycodestyle warnings
    "F",      # Pyflakes
    "I",      # isort
    "B",      # flake8-bugbear
    "C4",     # flake8-comprehensions
    "UP",     # pyupgrade
    "ARG",    # flake8-unused-arguments
    "SIM",    # flake8-simplify
]
ignore = [
    "E501",   # line too long (handled by formatter)
]

[tool.ruff.lint.isort]
known-first-party = ["my_project"]

[tool.ruff.format]
quote-style = "double"
indent-style = "space"

Commands

bash

# Check for issues
uv run ruff check .

# Fix auto-fixable issues
uv run ruff check --fix .

# Format code
uv run ruff format .

Type Checking with ty

Configuration

toml

[tool.ty]
python-version = "3.12"

Commands

bash

# Type check
uv run ty check

Project Structure

Single Package Project

text

my-project/
├── pyproject.toml
├── uv.lock
├── src/
│   └── my_project/
│       ├── __init__.py
│       └── main.py
└── tests/
    └── test_main.py

API + CLI with Shared Core

When building projects with both a web API and CLI:

text

your_app/
├── pyproject.toml
├── uv.lock
├── src/
│   └── your_app/
│       ├── __init__.py
│       ├── core/           # Shared business logic + DB access
│       │   ├── __init__.py
│       │   ├── models.py   # Domain models, Pydantic schemas
│       │   ├── services.py # Business logic
│       │   └── db.py       # Database access
│       ├── api/            # FastAPI endpoints import from core
│       │   ├── __init__.py
│       │   ├── main.py
│       │   └── routes/
│       └── cli/            # Click/Typer commands import from core
│           ├── __init__.py
│           └── main.py
└── tests/

Why this pattern:

•Single source of truth for business logic and validation
•No network dependency for CLI operations
•No code duplication between API and CLI
•CLI and API behave consistently
•Direct DB access means no latency penalty for CLI

Example Core Module

python

# src/your_app/core/services.py
from your_app.core.db import get_db
from your_app.core.models import User, CreateUserRequest

def create_user(request: CreateUserRequest) -> User:
    """Business logic shared by API and CLI."""
    db = get_db()
    user = User(name=request.name, email=request.email)
    db.add(user)
    db.commit()
    return user

Example API Using Core

python

# src/your_app/api/routes/users.py
from fastapi import APIRouter
from your_app.core.models import User, CreateUserRequest
from your_app.core.services import create_user

router = APIRouter()

@router.post("/users", response_model=User)
def create_user_endpoint(request: CreateUserRequest) -> User:
    return create_user(request)

Example CLI Using Core

python

# src/your_app/cli/main.py
import typer
from your_app.core.services import create_user
from your_app.core.models import CreateUserRequest

app = typer.Typer()

@app.command()
def add_user(name: str, email: str) -> None:
    """Create a new user."""
    request = CreateUserRequest(name=name, email=email)
    user = create_user(request)
    typer.echo(f"Created user: {user.id}")

Type Hints

Always Use Type Hints

python

# BAD
def process(data):
    return data["value"]

# GOOD
def process(data: dict[str, Any]) -> str:
    return data["value"]

Use Modern Syntax (3.10+)

python

# BAD - old style
from typing import List, Dict, Optional, Union

def process(items: List[str]) -> Optional[Dict[str, int]]:
    pass

# GOOD - modern syntax
def process(items: list[str]) -> dict[str, int] | None:
    pass

Pydantic for Data Validation

python

from pydantic import BaseModel, EmailStr

class CreateUserRequest(BaseModel):
    email: EmailStr
    name: str
    age: int | None = None

class User(BaseModel):
    id: int
    email: EmailStr
    name: str
    age: int | None = None

Naming Conventions

•snake_case: variables, functions, modules
•PascalCase: classes
•UPPER_SNAKE_CASE: constants

python

MAX_RETRIES = 3

class UserService:
    def get_user_by_id(self, user_id: int) -> User:
        pass

Imports

Order (handled by ruff)

•Standard library
•Third-party packages
•Local imports

python

import os
from pathlib import Path

from fastapi import FastAPI
from pydantic import BaseModel

from your_app.core.models import User
from your_app.core.services import create_user

Error Handling

Custom Exception Classes

python

class AppError(Exception):
    """Base application error."""
    def __init__(self, message: str, code: str) -> None:
        self.message = message
        self.code = code
        super().__init__(message)

class NotFoundError(AppError):
    """Resource not found."""
    def __init__(self, resource: str, id: str) -> None:
        super().__init__(
            message=f"{resource} with id {id} not found",
            code="NOT_FOUND"
        )

Result Pattern (Optional)

For functions that can fail predictably:

python

from dataclasses import dataclass
from typing import Generic, TypeVar

T = TypeVar("T")
E = TypeVar("E", bound=Exception)

@dataclass
class Ok(Generic[T]):
    value: T

@dataclass
class Err(Generic[E]):
    error: E

type Result[T, E] = Ok[T] | Err[E]

def parse_config(path: str) -> Result[Config, ConfigError]:
    try:
        data = load_file(path)
        return Ok(Config.model_validate(data))
    except ValidationError as e:
        return Err(ConfigError(str(e)))

CLI Applications

Required Stack

Purpose	Package
CLI framework	Typer
Colors/output	Rich
Spinners	halo
Progress bars	tqdm

Example CLI Setup

python

import typer
from rich import print
from rich.console import Console
from halo import Halo
from tqdm import tqdm

app = typer.Typer()
console = Console()

@app.command()
def process(
    path: str = typer.Argument(..., help="Path to process"),
    verbose: bool = typer.Option(False, "--verbose", "-v", help="Verbose output"),
) -> None:
    """Process files at the given path."""
    spinner = Halo(text="Loading files...", spinner="dots")
    spinner.start()

    try:
        files = load_files(path)
        spinner.succeed(f"Loaded {len(files)} files")
    except Exception as e:
        spinner.fail(f"Failed to load files: {e}")
        raise typer.Exit(1)

    for file in tqdm(files, desc="Processing"):
        process_file(file)

    print("[green]Done![/green]")

if __name__ == "__main__":
    app()

Dependencies

bash

uv add typer rich halo tqdm

LLM-Friendly Output

All CLIs must support both human and machine consumption:

python

import json
import typer
from rich import print
from rich.console import Console
from rich.table import Table
from pydantic import BaseModel

app = typer.Typer()
console = Console()

class User(BaseModel):
    id: str
    name: str
    email: str

@app.command()
def list_users(
    json_output: bool = typer.Option(False, "--json", help="Output as JSON for programmatic consumption"),
    limit: int = typer.Option(50, "--limit", "-l", help="Maximum number of users to return"),
) -> None:
    """
    List all users.

    Returns array of user objects with id, name, and email fields.
    Use --json for structured output suitable for piping to other tools or LLMs.
    """
    users = get_users(limit)

    if json_output:
        # Machine-readable: structured, no formatting
        print(json.dumps([u.model_dump() for u in users], indent=2))
    else:
        # Human-readable: colors, tables, pleasant
        table = Table(title="Users")
        table.add_column("Name", style="cyan")
        table.add_column("Email")
        for user in users:
            table.add_row(user.name, user.email)
        console.print(table)

if __name__ == "__main__":
    app()

Rules:

•--json flag on every command that outputs data
•JSON output: structured, complete, no ANSI codes (Rich auto-strips when not TTY)
•Default output: human-readable with Rich formatting
•Docstrings must explain what the command returns, not just what it does

Quick Reference

Tool	Purpose	Command
uv	Package management	`uv add`, `uv sync`, `uv run`
ruff	Linting + formatting	`ruff check`, `ruff format`
ty	Type checking	`ty check`

Pattern	Preference
Package manager	uv (not pip, poetry, pipenv)
Linter/formatter	ruff (replaces black, isort, flake8)
Type checker	ty
Type hints	Modern syntax (`list[str]` not `List[str]`)
Data validation	Pydantic
API framework	FastAPI
CLI framework	Typer (or Click)
Project structure	Shared core for API + CLI

Notes

•Always use uv for new projects
•ruff replaces multiple tools - don't install black, isort, or flake8 separately
•Use the shared core pattern when building API + CLI to avoid duplication
•Modern type hint syntax requires Python 3.10+
•Pydantic v2 for data validation and serialization