python-coding

Python Coder Standards

Apply these standards when writing, reviewing, or refactoring Python code. This provides the same expertise as the python-coder agent but applied directly in the current conversation context.

Quick Reference Checklist

Before providing any code or review, verify:

1. Absolute imports only
2. Modern type hints (X | None)
3. No prohibited practices (eval, exec, global, print, bare except, mutable defaults)
4. Pydantic validation at boundaries
5. No blocking I/O in async paths
6. asyncio.TaskGroup for structured concurrency
7. Google-style docstrings on public APIs
8. Structured logging (no f-strings in logger calls)
9. No assertions in production code
10. Immutable types preferred (tuple > list, frozenset > set)

See references/python-standards.md for full details on each item.

Detailed Standards

For the complete standards reference covering all patterns, examples, and anti-patterns, load:

references/python-standards.md

This reference covers:

• Design Principles - SOLID, dependency injection, composition over inheritance
• Asyncio & Concurrency - TaskGroup, firewall pattern, non-blocking I/O, synchronization
• Fault Tolerance - Exception isolation for independent tasks
• Error Handling - Exception hierarchy, no assertions in production
• Logging - Structured logging patterns
• Input Validation - Pydantic models at boundaries
• Naming & Immutability - PEP 8, no built-in shadowing, immutable defaults
• Imports & Type Hints - Absolute imports, modern syntax, Protocols
• Testing - TDD with pytest, async testing patterns
• Prohibited Practices - Complete list of banned patterns with alternatives

Output Guidelines

When writing code:

• Provide complete, runnable code with all imports
• Include type hints on all parameters and return values
• Add Google-style docstrings to public APIs
• Show example usage when helpful

When reviewing code:

• Reference specific violations from the standards
• Explain WHY each change is needed
• Provide corrected code
• Prioritize by severity