Skills

python

Installation
SKILL.md

Python Development Skill

Comprehensive guide to modern Python development covering conventions, testing, tooling, security, performance, and ecosystem best practices (2024-2025 standards).

Tier 2 Quick Start

For multi-file projects, team collaboration, and maintained code:

# Install Python 3.12 and uv (see Installation guides)
# Windows: winget install Python.Python.3.12 && winget install astral-sh.uv
# macOS: brew install python@3.13 uv
# Linux: apt install python3.13 && curl -LsSf https://astral.sh/uv/install.sh | sh

# Create new project with modern structure
uv init my-project
cd my-project

# Add dependencies
uv add requests httpx pydantic

# Add development dependencies
uv add --dev pytest pytest-cov ruff mypy

# Set up code quality (create pyproject.toml config - see assets/pyproject-toml-template.toml)
# Configure Ruff, mypy, pytest

# Write tests (pytest)
# tests/test_example.py - mirror src/ structure

# Run quality checks
uv run ruff check .          # Linting
uv run ruff format .         # Formatting
uv run mypy .                # Type checking
uv run pytest                # Run tests
uv run pytest --cov          # With coverage

When to Use This Skill

Invoke this skill when you need guidance on:

  • Installation & Setup: Installing Python 3.12/3.14, uv, setting up development environment
  • Code Style: Following PEP-8, configuring Ruff/Black, naming conventions
  • Project Structure: Organizing code with src/ layout, imports, package structure
  • Dependency Management: Using uv, Poetry, pip, virtual environments, pyproject.toml
  • Testing: Writing pytest tests, fixtures, parametrization, coverage, mocking
  • Type Hints: Modern typing patterns, mypy configuration, protocols, generics
  • Code Quality: Setting up Ruff, mypy, Bandit, pre-commit hooks, CI/CD
  • Async Programming: asyncio patterns, async/await, TaskGroup, structured concurrency
  • Security: OWASP best practices, input validation, dependency scanning
  • Performance: Profiling, optimization patterns, Cython
  • Packaging: pyproject.toml, PyPI publishing, versioning, distribution
  • Common Libraries: Standard library essentials, ecosystem overview
  • Documentation: Docstrings (PEP-257), Sphinx, documentation generation

Overview

This skill provides modern Python development guidance aligned with current best practices (2024-2025):

Core Standards:

  • Style: PEP-8 (enforced via Ruff)
  • Type Hints: PEP-484, PEP-526, modern syntax (PEP-604: str | None)
  • Packaging: PEP-517, PEP-518 (pyproject.toml)
  • Docstrings: PEP-257
  • Testing: pytest (industry standard)
  • Dependency Management: uv (fastest) or Poetry (feature-rich)

Modern Tooling (2024-2025):

  • Linter/Formatter: Ruff (replaces flake8, isort, optionally Black)
  • Type Checker: mypy with strict mode
  • Test Framework: pytest
  • Dependency Manager: uv or Poetry
  • Security Scanner: Bandit + pip-audit
  • Project Config: pyproject.toml (universal)

Project Structure:

  • src/ layout (modern standard, not flat layout)
  • Absolute imports preferred over relative
  • Tests mirror src/ structure
  • pyproject.toml for all configuration

Official Sources: All guidance backed by official Python documentation, PEPs, and tool documentation:

  • docs.python.org
  • peps.python.org
  • docs.astral.sh/ruff
  • docs.astral.sh/uv
  • mypy-lang.org
  • docs.pytest.org

Quick Tier Selection

Choose your complexity level:

🎯 Tier 1: Minimal (Simple Scripts)

→ Single-file utilities, converting scripts, one-off automation → Just Python code - no tooling overhead → Jump to Minimal Guidance

📦 Tier 2: Standard (Organized Projects)

→ Multi-file modules, team projects, maintained code → Modern project structure + testing → Jump to Standard Guidance

🚀 Tier 3: Full (Production Systems)

→ PyPI packages, enterprise systems, production deployments → Complete tooling ecosystem → Jump to Full Guidance

Not sure? Default to Tier 2 (Standard) - it covers most use cases.

Tier 1: Minimal (Simple Scripts)

For: Single-file utilities, script conversions, and simple automation.

Setup

Just Python 3.12+ - no additional tooling required:

  • ✅ No uv, no virtual environment needed
  • ✅ No pyproject.toml, no src/ layout
  • ✅ Optional: Install Ruff for quick linting (pip install ruff)

Code Standards

Follow these simple principles:

  • PEP-8 naming: snake_case for functions/variables, PascalCase for classes
  • Type hints: Add for clarity (helps readers understand your code)
  • Docstrings: Simple descriptions of what functions do
  • pathlib: Use pathlib.Path for file operations (not os.path)
  • Built-ins: Use Python's built-in json, datetime, sys modules

Example: Simple Logging Script

#!/usr/bin/env python3
"""Simple logging utility - converts event to JSON."""

import json
import sys
from datetime import datetime, timezone
from pathlib import Path


def log_event(event_name: str, data: dict) -> None:
    """
    Log an event to daily JSONL file.

    Args:
        event_name: Name of the event
        data: Event data to log
    """
    log_dir = Path(__file__).parent / "logs"
    log_dir.mkdir(exist_ok=True)

    now = datetime.now(timezone.utc)
    log_file = log_dir / f"{now:%Y-%m-%d}.jsonl"

    entry = {
        "timestamp": now.isoformat(),
        "event": event_name,
        "data": data
    }

    with open(log_file, "a", encoding="utf-8") as f:
        f.write(json.dumps(entry) + "\n")


if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: script.py <event_name>", file=sys.stderr)
        sys.exit(1)
    event_name = sys.argv[1]
    data = json.loads(sys.stdin.read())
    log_event(event_name, data)

Optional: Quick Linting

If you want to check your code style:

# Install Ruff (optional)
pip install ruff

# Check code
ruff check script.py

# Format code
ruff format script.py

When to Upgrade to Tier 2

Consider upgrading to Tier 2 (Standard) when:

  • ✅ Script grows to 3+ files
  • ✅ Multiple people working on code
  • ✅ Need automated testing
  • ✅ Managing external dependencies
  • ✅ Code will be maintained long-term

Tier 2: Standard (Organized Projects)

For: Multi-file modules, team projects, and maintained code.

Modern Python project setup with proper structure, testing, and quality tools.

Installation

Install Python 3.12+ and uv (modern dependency manager). See platform-specific guides:

Project Setup and Organization

Conventions and Style - Follow PEP-8 with Ruff for linting and formatting:

Project Structure - Use modern src/ layout for proper packaging:

Dependency Management - Choose between uv (fastest), Poetry (feature-rich), or pip+venv:

Testing and Quality

Testing - Use pytest with fixtures, parametrization, and coverage:

Type Hints - Modern typing with mypy, protocols, and generics:

Code Quality - Set up Ruff, mypy, Bandit, and pre-commit hooks:

Advanced Development

Async Programming - asyncio patterns, TaskGroup, structured concurrency:

Security - OWASP best practices, input validation, secrets management:

Performance - Profiling, optimization patterns, memory efficiency:

Packaging - Publish to PyPI with pyproject.toml and semantic versioning:

Libraries - Standard library essentials and ecosystem overview:

Documentation - PEP-257 docstrings, Sphinx, ReadTheDocs:

Tier 3: Full (Production Systems)

For: PyPI packages, enterprise systems, and production deployments.

Everything in Tier 2 (Standard) PLUS:

Security & Quality

  • Security scanning: Bandit for code security, pip-audit for dependency vulnerabilities
  • Pre-commit hooks: Automated checks before every commit
  • Comprehensive type coverage: mypy strict mode across entire codebase
  • Code coverage enforcement: Minimum coverage requirements in CI/CD

Distribution

  • PyPI packaging: Build and publish Python packages to PyPI
  • Semantic versioning: Follow semver for version management
  • Binary distributions: Create wheels for faster installation
  • Multi-platform support: Test and build for Windows/macOS/Linux

CI/CD

  • GitHub Actions / GitLab CI: Automated testing on every push
  • Multi-version testing: Test against Python 3.12, 3.13, 3.14
  • Coverage tracking: Automated coverage reports and enforcement
  • Release automation: Automatic PyPI publishing on tags

Documentation

  • Sphinx: Generate API documentation from docstrings
  • ReadTheDocs: Host documentation with versioning
  • Comprehensive docstrings: Google or NumPy style for all public APIs
  • Examples and tutorials: User-facing documentation with examples

Detailed Guides

See these reference files for comprehensive production guidance:

Reference Files

All detailed guidance is in the references/ directory:

Installation:

Core Development:

Advanced Topics:

Assets

Template Files:

Testing and Evaluation

This skill includes formal evaluation scenarios to validate effectiveness:

  • See Skill Evaluation Scenarios for test cases covering:
    • Minimal Tier (simple scripts)
    • Standard Tier (new projects, code quality tools)
    • Full Tier (production systems)
    • Alternative workflows (Poetry vs uv)
    • Success criteria and testing methodology

Related Skills

  • git-commit: Git commit workflow (complements Python project workflow)
  • git:git-config: Git configuration (Python projects use Git)
  • code-quality:markdown-linting: Markdown linting via plugin (Python projects have READMEs)

Version History

  • 1.1.2 (2025-11-25): Comprehensive improvements
    • Fixed deprecated datetime.utcnow() in example code (use datetime.now(timezone.utc))
    • Added Python 3.13 coverage (was missing between 3.12 and 3.14)
    • Updated PEP URLs to modern format (peps.python.org instead of python.org/dev/peps/)
    • Added UTF-8 encoding to file open in example code
    • Added error handling for missing CLI arguments in example
  • 1.1.1 (2025-11-25): Version audit update
    • Updated tool version references (Ruff 0.14.x, mypy 1.18.x, pytest 9.x)
    • Verified all reference file links exist and are correct
    • Validated YAML frontmatter against official requirements
  • 1.1.0 (2025-11-18): Adaptive tiered guidance
    • Added three-tier complexity system (Minimal/Standard/Full)
    • Tier 1: Minimal guidance for simple scripts (no tooling overhead)
    • Tier 2: Standard guidance for organized projects (original Quick Start)
    • Tier 3: Full guidance for production systems (comprehensive tooling)
    • Context-aware skill automatically selects appropriate tier based on query
    • Prevents "big bang" overengineering for simple scenarios
  • 1.0.0 (2025-11-17): Initial release
    • Comprehensive modern Python guidance (2024-2025 standards)
    • Migrated installation docs from docs/python/
    • Added 12 detailed reference files
    • Modern tooling: Ruff, uv, pytest, mypy
    • pyproject.toml template

Official Documentation

Python:

Tools:

PEPs (Python Enhancement Proposals):

Last Updated

Date: 2025-11-28 Model: claude-opus-4-5-20251101

Related skills

More from melodic-software/claude-code-plugins

Installs
1
Repository
melodic-softwar…-plugins
GitHub Stars
63
First Seen
Apr 18, 2026
Security Audits
Gen Agent Trust HubPass
SocketPass
SnykPass