- Home
- Skills
- Anilcancakir
- My Claude Code
- Laravel Infrastructure
laravel-infrastructure_skill
4
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill anilcancakir/my-claude-code --skill laravel-infrastructure- SKILL.md6.2 KB
Overview
This skill provides production-ready patterns and operational guidance for Laravel infrastructure: Horizon queues, Octane performance, Reverb WebSockets, Redis caching, and PostgreSQL. It focuses on reliable queue management, real-time broadcasting, memory-safe Octane usage, cache and session reliability, and database performance techniques. Activate when working with jobs, workers, caching, broadcasting, or database tuning.
How this skill works
The skill inspects queue and worker definitions, Horizon supervisor configuration, and common queue failure modes to diagnose stuck jobs or worker crashes. It checks Octane-safe coding patterns, singleton flush lists, and exposes memory-leak risks. It validates Reverb broadcast configuration and WebSocket connectivity, and verifies Redis caching, locks, and session settings. It also highlights PostgreSQL-specific schema and query patterns for JSONB, full-text, and slow-query detection.
When to use it
- When a job fails or repeatedly retries
- When queues are stuck or workers stopped
- When cache entries won’t clear or Redis connections fail
- When broadcast or WebSocket events don’t arrive in real time
- When Octane shows memory growth or unexpected behavior
- When PostgreSQL shows slow queries, N+1 issues, or connection timeouts
Best practices
- Define Horizon supervisors per environment and limit memory/tries to prevent runaway processes
- Use Octane-safe patterns: avoid request-accumulating statics and flush singletons between requests
- Use Redis locks for idempotent job processing and Cache::tags for targeted invalidation
- Run Horizon under Supervisor in production with long stopwaitsecs and proper user permissions
- Broadcast via Reverb with authenticated channels and test with Laravel Echo in staging
- Use JSONB indices, full-text search, and eager loading to prevent N+1 and speed queries
Example use cases
- Diagnose a stalled Horizon queue and restart supervisors with safe job backoff
- Harden Octane by identifying singleton services to flush and reducing worker counts to test memory behavior
- Fix missing real-time updates by validating Reverb env vars, ports, and Laravel Echo listeners
- Implement Redis tag-based cache invalidation to avoid full cache flushes in production
- Tune PostgreSQL queries: add JSONB GIN indices, convert slow Eloquent loops to batched queries
FAQ
Check worker processes, ensure Supervisor is running, review job timeouts and memory limits, and inspect Redis for connection errors. Restart Horizon after addressing root cause.
How do I find Octane memory leaks?
Search for static properties, long-lived singletons, or in-memory accumulators. Add classes to Octane flush config and prefer request-scoped services or external caches.