How does this skill prevent GenServer bottlenecks?

It applies the 'Iron Law' which states GenServers are bottlenecks by design; the skill suggests using ETS for high-concurrency reads while maintaining serialized writes.

When should I use Oban versus Broadway according to this skill?

The skill guides you to use Oban for background jobs requiring database persistence and Broadway for external queue data ingestion (like SQS or Kafka) with batching.

Does it support distributed Elixir applications?

Yes, it provides a comparison between local Registry and cluster-wide :pg for process groups, as well as guidance on using Horde for distributed supervisors.

Can this skill help with Elixir error handling?

Yes, it provides strategies for Supervision trees (one_for_one, rest_for_one) and explains when to use Task.Supervisor to ensure your caller process doesn't crash.

Elixir OTP Architect

Name: Elixir OTP Architect
Author: georgeguimaraes

bygeorgeguimaraes

•

API Development

Guides Elixir developers in implementing robust, concurrent systems using Open Telecom Platform (OTP) best practices and design patterns.

This skill provides deep expertise in Elixir's OTP ecosystem, helping developers navigate complex concurrency, state management, and fault-tolerance requirements. It enforces the 'Iron Law' of GenServers to prevent bottlenecks, aids in choosing between background job tools like Oban and Broadway, and provides architectural decision trees for supervision trees, task management, and distributed process registries. Whether you are building a real-time caching layer or a high-throughput data pipeline, this skill ensures your Elixir application is performant, scalable, and follows idiomatic Elixir patterns.

Key Features

01Supervision tree strategy design for fault-tolerant systems

0276 GitHub stars

03Task.Supervisor vs Task.async implementation guidance

04Advanced GenServer bottleneck analysis and ETS optimization patterns

05Decision support for choosing between Oban (jobs) and Broadway (streams)

06Distributed process management using Registry, PartitionSupervisor, and :pg

Use Cases

01Designing a fault-tolerant background processing system for external APIs

02Optimizing a slow GenServer by implementing ETS read-concurrency

03Architecting dynamic process management for millions of concurrent users

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add georgeguimaraes/claude-code-elixir otp-thinking

For use in Claude.ai and ChatGPT

Download Skill