Resilience Patterns

•
69
•

Implements production-grade fault tolerance and reliability patterns for distributed systems and LLM integrations.

The Resilience Patterns skill provides a comprehensive toolkit for building reliable, fault-tolerant applications using industry-standard strategies like circuit breakers, bulkhead isolation, and intelligent retry mechanisms. Designed specifically for the complexities of modern distributed systems and LLM-based workflows, it helps developers prevent cascading failures, manage API rate limits with exponential backoff, and implement graceful degradation through fallback chains. Whether you are building high-traffic backends or complex multi-agent AI systems, this skill offers the production-ready code templates and architectural guidance needed to ensure system stability under stress.

Key Features

01Bulkhead isolation patterns for resource partitioning and failure containment
02Circuit breaker implementation to prevent system-wide cascading failures
03Advanced retry strategies featuring exponential backoff and randomized jitter
04Token budget management and context truncation strategies for AI APIs
0569 GitHub stars
06LLM-specific fallback chains for multi-model reliability and cost control

Use Cases

01Protecting backend microservices from being overwhelmed by failing downstream dependencies
02Implementing production-ready error handling for OpenAI, Anthropic, or custom LLM integrations
03Building reliable multi-agent AI systems that handle API downtime gracefully