How does the Circuit Breaker pattern work here?

It monitors for failure thresholds in external calls and 'trips' the circuit to prevent cascading failures, allowing the system to fail fast and recover gracefully once the service is stable.

What is Bulkhead isolation?

It's a pattern that partitions system resources into isolated pools, ensuring that a failure or slowdown in one service component doesn't consume all system resources and crash the entire application.

What are Resilience Patterns in Claude Code?

They are standardized code implementations and logic flows used to make applications more robust against failures in distributed environments or external APIs.

Does this skill support LLM-specific failures?

Yes, it includes specific patterns for LLM fallbacks (switching models on failure), token budget management, and handling model-specific errors like rate limits or context overflows.

Are the retry strategies configurable?

Yes, the skill provides templates for configurable retry logic including custom backoff multipliers, jitter to prevent 'thundering herd' problems, and specific error classification.

System Resilience Patterns

Name: System Resilience Patterns
Author: yonatangross

byyonatangross

•

API Development

Implements production-grade fault tolerance patterns like circuit breakers, bulkheads, and retries for robust distributed systems and LLM integrations.

About

The Resilience Patterns skill provides a comprehensive framework for building fault-tolerant distributed systems and reliable LLM-based workflows. It equips developers with standardized implementations for circuit breakers to prevent cascading failures, bulkhead isolation to partition resources, and intelligent retry strategies with exponential backoff. Specifically optimized for AI development, it also includes patterns for LLM API fallback chains and token budget management, ensuring that production services remain stable and responsive even when external dependencies or AI models encounter transient errors or outages.

Key Features

Standardized Circuit Breaker implementation to prevent cascading system failures
Token budget guards for managing context limits and API costs
29 GitHub stars
Bulkhead isolation strategies to partition resources and limit blast radii
Intelligent retry logic featuring exponential backoff and jitter
LLM-specific fallback chains for multi-model reliability and high availability

Use Cases

Managing rate limits and transient network errors in production-scale LLM integrations
Building high-availability backend services that depend on multiple third-party APIs
Designing multi-agent AI systems that require graceful degradation when primary models fail

About

Key Features

Standardized Circuit Breaker implementation to prevent cascading system failures
Token budget guards for managing context limits and API costs
29 GitHub stars
Bulkhead isolation strategies to partition resources and limit blast radii
Intelligent retry logic featuring exponential backoff and jitter
LLM-specific fallback chains for multi-model reliability and high availability

Use Cases

Managing rate limits and transient network errors in production-scale LLM integrations
Building high-availability backend services that depend on multiple third-party APIs
Designing multi-agent AI systems that require graceful degradation when primary models fail