Table of Contents

👉 Try the demo: https://github.com/LM-Kit/lm-kit-net-samples/tree/main/console_net/ai-agents/resilience/agent_resilience

Agent Resilience Policies Lab for C# .NET Applications


🎯 Purpose of the Demo

An interactive console app that demonstrates the four core resilience policies on a controllable flaky operation. Pick a policy, choose the failures-before-success profile, watch the policy behave. The same policies wrap real agent calls in production via AgentBuilder.WithFilters(...) and middleware.

No model is loaded; the demo focuses on policy mechanics.


👥 Industry Target Audience

  • Production teams hardening agent pipelines for real traffic.
  • SRE / ops: understand state transitions before they ship.
  • App developers learning the declarative resilience surface area.
  • Architects picking the right policy or combination for a workflow.

🚀 Problem Solved

Production agents fail under load. Retry covers transient errors. Circuit-breaker stops the bleed when a downstream is dead. Fallback returns a safe default. Composite chains them. The lab makes each behavior visible on a known-failure profile, so you can pick the right combination for your code path.


💻 Application Overview

Interactive menu (no command-line arguments) with five modes:

Mode What it does
Retry Run RetryPolicy with exponential backoff + jitter; pick failures-before-success interactively.
Breaker Run CircuitBreakerPolicy against an always-failing operation; watch state transitions.
Fallback Run FallbackPolicy<T> with a default return value.
Composite Run CompositePolicy wrapping Retry + Fallback.
All Run every demo back-to-back.
Quit Exit.

✨ Key Features

  • RetryPolicy with WithExponentialBackoff, WithJitter, OnRetry.
  • CircuitBreakerPolicy with OnStateChange.
  • FallbackPolicy<T> with shouldHandle predicate.
  • CompositePolicy chained via Wrap(...).

🧠 Model

  • None. No inference is required; the demo is pure policy mechanics.

🛠️ Getting Started

📋 Prerequisites

  • .NET 8.0 or later

▶️ Running the Application

git clone https://github.com/LM-Kit/lm-kit-net-samples
cd lm-kit-net-samples/console_net/ai-agents/resilience/agent_resilience
dotnet run

Pick a policy from the menu.

📚 Related Content

Share