Concurrency

Overview

Concurrency enables programs to handle multiple tasks efficiently. This skill covers async/await patterns across Rust (tokio), Python (asyncio), TypeScript (Promises), and Go (goroutines). Includes parallelism strategies, race condition prevention, deadlock handling, thread safety patterns, channel-based communication, and work queue implementations.

Agent Specializations

When implementing concurrency, delegate to the appropriate specialist:

1. senior-software-engineer (Opus) - DEFAULT. Architectural decisions, threading models, message-passing vs shared-state, implementing concurrent code, async functions, worker pools, rate limiters, identifying race conditions, TOCTOU vulnerabilities, lock ordering, distributed concurrency, consistency models, distributed locks, saga patterns
2. software-engineer (Sonnet) - ONLY for unit tests or boilerplate async handlers following established patterns

Instructions

1. Rust Async/Await with Tokio