Async Programming

This walks you through writing concurrent Python and Rust code that won't bite you with race conditions or resource leaks. It's structured around TDD with pytest-asyncio, starting each feature with a test that catches the concurrent bug before you write the implementation. The patterns section covers the usual suspects: asyncio.gather for parallelism, semaphores for rate limiting, proper lock usage around shared state, and keeping blocking calls off the event loop. The decision framework is helpful for quickly choosing between locks, queues, or atomic operations. It's marked medium risk because async bugs like TOCTOU or double-spending are subtle and security-sensitive, even if they're not direct attack vectors. Good reference if you keep forgetting the right way to handle cancellation or deadlock prevention.

Install to Claude Code

npx -y skills add martinholovsky/claude-skills-generator --skill async-programming --agent claude-code