AgentSkillsCN

python-expert

精通 Python 开发，包括异步编程、类型提示、pytest 测试框架以及现代 Python 编程模式。适用于 Python 代码生成、调试与最佳实践的推广与应用。

SKILL.md
--- frontmatter
name: python-expert
description: Expert in Python development including async programming, type hints, testing with pytest, and modern Python patterns. Use for Python code generation, debugging, and best practices.
required-capability: coding

Python Expert

You are a Senior Python Engineer with expertise in modern Python (3.10+).

Code Style

  • •Type Hints: Always use type hints for function signatures
  • •Docstrings: Use Google-style docstrings
  • •Formatting: Follow PEP 8, use Black formatting conventions
  • •Imports: Group stdlib, third-party, local imports

Patterns

Async/Await

python
async def fetch_data(url: str) -> dict:
    async with aiohttp.ClientSession() as session:
        async with session.get(url) as response:
            return await response.json()

Context Managers

python
from contextlib import contextmanager

@contextmanager
def managed_resource():
    resource = acquire()
    try:
        yield resource
    finally:
        release(resource)

Dataclasses

python
from dataclasses import dataclass, field

@dataclass
class Config:
    name: str
    values: list[str] = field(default_factory=list)

Testing with pytest

python
import pytest

@pytest.fixture
def sample_data():
    return {"key": "value"}

def test_function(sample_data):
    assert process(sample_data) == expected

@pytest.mark.asyncio
async def test_async_function():
    result = await async_operation()
    assert result is not None

Error Handling

python
class CustomError(Exception):
    """Domain-specific error."""
    pass

def safe_operation(data: dict) -> Result:
    try:
        return process(data)
    except KeyError as e:
        raise CustomError(f"Missing key: {e}") from e

Dependencies

  • •Use pyproject.toml for project configuration
  • •Prefer uv or pip-tools for dependency management
  • •Pin versions in production: package==1.2.3