| name | go-concurrency |
| description | Use this skill when writing Go goroutines, channels, mutexes, sync primitives, context cancellation, or any concurrent Go code to avoid data races and goroutine leaks. |
Go Concurrency
Critical patterns for safe and correct concurrent Go code. Covers goroutine lifecycle, data races, channels vs mutexes, context propagation, sync package usage, and goroutine leak prevention.
When to Use This Skill
- Writing goroutines and managing their lifecycle
- Choosing between channels and mutexes
- Using context for cancellation and propagation
- Working with sync.WaitGroup, sync.Mutex, sync.Cond
- Using errgroup for goroutine error handling
- Detecting and fixing data races
How It Works
- Identifies critical concurrency anti-patterns
- Explains data races, deadlocks, and goroutine leaks
- Shows safe patterns for common concurrent workloads
- Covers both foundational concepts and practical implementation
Categories Covered
- Concurrency Foundations — goroutines vs parallelism, workload types, channel vs mutex, context
- Concurrency Practice — goroutine leaks, data races, channels, sync primitives, errgroup
Present Results to User
When identifying issues:
❌ Found: Goroutine with no exit condition (leak)
📍 Location: services/worker.go:78
⚠️ Impact: CRITICAL - Memory grows unbounded in production
✅ Fix: Use context cancellation or done channel
[Show incorrect code]
[Show correct code]
[Explain why the fix matters]
Troubleshooting
Channel vs mutex:
- Use channels when transferring ownership of data between goroutines
- Use mutexes when protecting shared state accessed by multiple goroutines
Race detector:
- Always run
go test -race — races that seem harmless in testing cause crashes in production
