Python Project Setup

Python Project Setup Skill

A specialized skill for setting up modern Python projects using uv tooling, best practices, and standardized project structures.

Overview

This skill enables Claude to autonomously set up Python projects following modern best practices. It promotes the use of uv (the fast Python package manager), enforces PEP8 standards, type hints, and proper project structure with src layout.

Capabilities

When activated, this skill provides:

1. Modern Tooling with uv

   • Use uv instead of pip/python directly
   • Initialize virtual environments: uv venv
   • Manage dependencies: uv add, uv remove
   • Sync dependencies: uv sync
   • Run scripts: uv run

2. Project Structure Detection & Setup

   • Detect if project is new or existing
   • Analyze current structure and adapt recommendations
   • Promote src/package_name layout
   • Create proper __init__.py files
   • Setup __main__.py entry points

3. pyproject.toml Management

   • Generate modern pyproject.toml
   • Configure project metadata
   • Setup dependencies and dev dependencies
   • Include tool configurations (ruff, mypy, pytest)
   • Define entry points for CLI applications

4. Library Selection

   • Present interactive menu for library choices
   • CLI applications: click (commands), rich (output), inquirer (interactive prompts)
   • TUI applications: textual (framework), rich (rendering)
   • Pure libraries: minimal dependencies
   • Smart recommendations based on project type

5. Standards Enforcement

   • PEP8 compliance via ruff
   • Type hints for all functions/methods
   • Proper docstring formats (Google/NumPy style)
   • Follow PEP8 and modern Python standards

Usage

This skill activates automatically when:

• User requests: "Create a new Python project"
• User asks: "Setup Python project structure"
• User mentions: "Initialize Python package"
• User wants to: "Migrate to uv"
• Claude detects Python project initialization
• pyproject.toml creation or modification is discussed

Automatic Defaults:

• Uses current directory as project root
• Derives package name from directory name
• Detects and uses current Python version
• No tests (user can add later)
• Minimal dependencies by default

Only ONE Question:

• Project type + libraries combined (default: minimal)

Project Setup Approach

For New Projects

IMPORTANT: Always work in the current directory as the project root.

1. Gather Requirements

   • Package name (derive from current directory name by default)
   • Ask ONE merged question: Project type + libraries combined:

     What type of project? (press Enter for CLI with click+rich)
     [1] CLI with click+rich (default)
     [2] CLI with argparse
     [3] CLI with typer
     [4] TUI with textual
     [5] Generic Python Project
     

   • Don't ask about tests - skip by default

2. Assume Current Directory & Python Version

   • Use current folder as project root (don't create new directory)
   • Derive package name from current folder name (convert to valid Python identifier)
   • Example: my-awesome-app/ → package name: my_awesome_app
   • Detect current Python version and use as project requirement
   • Example: python --version → Python 3.11.5 → requires-python = ">=3.11"

3. Check Environment

   # Detect current Python version
   python --version  # Use this as the project's Python requirement
   
   # Verify uv is installed
   uv --version
   
   # If not installed, provide instructions:
   # macOS/Linux: curl -LsSf https://astral.sh/uv/install.sh | sh
   # Windows: powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
   

4. Create Minimal Project Structure in Current Directory

   ./                           # Current directory (project root)
   ├── src/
   │   └── package_name/
   │       ├── __init__.py
   │       └── __main__.py      # Entry point
   ├── pyproject.toml           # Minimal config
   ├── README.md
   └── .gitignore
   

   Note: No tests/, no .python-version - keep it minimal!

5. Skip Tests by Default

   • Do NOT ask about tests
   • Do NOT create tests/ directory
   • Do NOT include pytest in dev dependencies
   • User can add tests later if needed

6. Generate Minimal pyproject.toml

   [project]
   name = "package-name"
   version = "0.1.0"
   description = "Project description"
   authors = [
       {name = "Author Name", email = "author@example.com"}
   ]
   readme = "README.md"
   requires-python = ">=3.11"  # Use detected Python version (e.g., 3.11, 3.12, etc.)
   dependencies = [
       # Based on library selection
   ]
   
   [project.optional-dependencies]
   dev = [
       "ruff>=0.1.0",
       "mypy>=1.8.0",
   ]
   # Note: No pytest by default - add when needed
   
   [project.scripts]
   package-name = "package_name.__main__:main"
   
   [build-system]
   requires = ["hatchling"]
   build-backend = "hatchling.build"
   
   [tool.ruff]
   line-length = 88
   target-version = "py311"
   select = ["E", "F", "I", "N", "W", "UP"]
   
   [tool.mypy]
   python_version = "3.11"
   strict = true
   warn_return_any = true
   warn_unused_configs = true
   
   [tool.pytest.ini_options]
   testpaths = ["tests"]
   python_files = ["test_*.py"]
   

7. Initialize with uv in Current Directory

   # Already in project directory (current folder)
   
   # Create virtual environment
   uv venv
   
   # Sync dependencies from pyproject.toml
   uv sync
   
   # Add project in editable mode
   uv pip install -e .
   
   # Add development dependencies (includes pytest if tests opted in)
   uv sync --group dev
   

8. Create Entry Point Examples

   For src/package_name/__main__.py:

   """Main entry point for package_name."""
   
   def main() -> None:
       """Execute the main program."""
       print("Hello from package_name!")
   
   if __name__ == "__main__":
       main()
   

   For CLI apps with click:

   """Main entry point for package_name CLI."""
   
   import click
   from rich.console import Console
   
   console = Console()
   
   @click.group()
   @click.version_option()
   def main() -> None:
       """Package name CLI application."""
       pass
   
   @main.command()
   @click.option("--name", default="World", help="Name to greet")
   def hello(name: str) -> None:
       """Greet someone."""
       console.print(f"[bold green]Hello, {name}![/bold green]")
   
   if __name__ == "__main__":
       main()
   

For Existing Projects

IMPORTANT: Always work in the current directory, don't create subdirectories.

1. Analyze Current Directory Structure

   • Check for setup.py, requirements.txt, or pyproject.toml
   • Identify current package structure
   • Assess if src layout is used
   • Detect existing tests/ directory

2. Detect Context

   • Is uv already being used?
   • What's the current dependency management approach?
   • Are there existing entry points?
   • Are tests already present? (skip test skeleton question if yes)

3. Skip Tests Question

   • Do NOT ask about tests for existing projects
   • User can add tests manually if needed

4. Provide Migration Path

   # Already in project directory (current folder)
   
   # Convert requirements.txt to pyproject.toml
   # Create pyproject.toml with [project.dependencies]
   
   # Initialize uv for existing project
   uv venv
   
   # Import existing requirements
   uv pip install -r requirements.txt
   
   # Generate lock file
   uv sync
   

5. Recommend Structure Improvements

   • Suggest moving to src layout if not present
   • Identify missing init.py files
   • Recommend adding main.py if needed
   • Suggest adding type hints if missing

6. Adapt, Don't Force

   • Respect existing conventions if reasonable
   • Provide gradual migration suggestions
   • Explain benefits of each recommendation

Output Format

Provide setup results in this format:

## Python Project Setup Complete

### Project Directory
**Working in**: `./` (current directory)
**Package name**: `package_name`

### Project Structure
[Show created directory tree relative to current directory]

### Configuration
- **Python version**: [Detected from current environment, e.g., 3.11+]
- **Package manager**: uv
- **Layout**: src layout
- **Entry point**: src/package_name/__main__.py
- **Tests**: Not included (add later if needed)

### Dependencies Installed
**Main dependencies:**
- [list dependencies]

**Development dependencies:**
- ruff (linting)
- mypy (type checking)

### Next Steps
1. Activate virtual environment:
   \`\`\`bash
   source .venv/bin/activate  # Linux/macOS
   .venv\\Scripts\\activate     # Windows
   \`\`\`

2. Run your application:
   \`\`\`bash
   uv run python -m package_name
   # or
   uv run package-name
   \`\`\`

3. Add new dependencies:
   \`\`\`bash
   uv add package-name
   \`\`\`

4. Lint code:
   \`\`\`bash
   uv run ruff check .
   uv run mypy src
   \`\`\`

### Development Commands
- Add dependency: `uv add package-name`
- Add dev dependency: `uv add --dev package-name`
- Remove dependency: `uv remove package-name`
- Update dependencies: `uv sync`
- Run script: `uv run python script.py`

Standards Enforced

Always Use uv Commands

• ❌ pip install package

• ✅ uv add package

• ❌ pip install -r requirements.txt

• ✅ uv sync

• ❌ python script.py

• ✅ uv run python script.py

• ❌ python -m venv .venv

• ✅ uv venv

Project Structure Standards

src/package_name/          # Use src layout
├── __init__.py           # Package marker (can be empty or contain version)
├── __main__.py           # Entry point with main() function
├── module1.py            # Application modules
└── module2.py

Code Standards

• All functions must have type hints
• Use """Docstrings""" for all public functions/classes
• Follow PEP8 (enforced by ruff)
• Line length: 88 characters (Black standard)
• Use absolute imports from package root

Example Type-Hinted Code

from typing import Optional

def process_data(
    input_data: list[str],
    max_items: int = 10,
    filter_empty: bool = True
) -> dict[str, int]:
    """Process input data and return statistics.

    Args:
        input_data: List of strings to process
        max_items: Maximum number of items to process
        filter_empty: Whether to filter out empty strings

    Returns:
        Dictionary with processing statistics
    """
    result: dict[str, int] = {}
    # Implementation
    return result

Library Recommendations

For CLI Applications

dependencies = [
    "click>=8.1.0",      # Command-line interface framework
    "rich>=13.7.0",      # Beautiful terminal output
    "inquirer>=3.2.0",   # Interactive prompts
]

For TUI Applications

dependencies = [
    "textual>=0.47.0",   # Modern TUI framework
    "rich>=13.7.0",      # Rendering engine
]

For Libraries

dependencies = [
    # Only essential dependencies
    # Keep it minimal
]

Best Practices

1. Work in current directory - Don't create nested project folders
2. Use current Python version - Detect from environment, don't ask
3. Ask one merged question - Project type + libraries combined
4. Start with minimal dependencies - Add more as needed
5. Always include type hints - Helps with IDE support and catches bugs
6. Use src layout - Prevents import issues and packaging problems
7. Pin major versions only - e.g., >=8.1.0 not ==8.1.0
8. Separate dev dependencies - Use [project.optional-dependencies]
9. Document with docstrings - Every public function and class
10. Don't ask about tests - Skip by default, users add when ready
11. Use ruff for linting - Fast, comprehensive, replaces flake8/isort/black

Common Patterns

CLI Application with Click + Rich

import click
from rich.console import Console
from rich.table import Table

console = Console()

@click.command()
@click.argument("name")
@click.option("--count", default=1, help="Number of greetings")
def greet(name: str, count: int) -> None:
    """Greet NAME COUNT times."""
    for _ in range(count):
        console.print(f"[bold blue]Hello, {name}![/bold blue]")

TUI Application with Textual

from textual.app import App, ComposeResult
from textual.widgets import Header, Footer, Button

class MyApp(App):
    """A simple Textual application."""

    def compose(self) -> ComposeResult:
        yield Header()
        yield Button("Click me!", id="btn")
        yield Footer()

    def on_button_pressed(self, event: Button.Pressed) -> None:
        """Handle button press."""
        self.notify("Button clicked!")

if __name__ == "__main__":
    app = MyApp()
    app.run()

Integration

This skill works seamlessly with:

• New Python project creation
• Existing project modernization
• Migration from pip to uv
• Package structure refactoring
• Dependency management tasks

Converted and distributed by TomeVault — claim your Tome and manage your conversions.