architecture▌

Build sophisticated LLM applications with LangChain 1.x and LangGraph for agents, memory, and tool integration. \n \n LangGraph provides the standard agent framework with StateGraph for explicit state management, durable execution, human-in-the-loop inspection, and checkpointing across sessions \n Supports ReAct agents, plan-and-execute workflows, multi-agent supervision, and structured tool invocation with Pydantic schemas \n Memory systems include ConversationBufferMemory, ConversationSummaryM

<why_now>

Serverless architecture enables building complete applications without managing servers. Design event-driven, scalable systems using managed compute services, databases, and messaging systems. Pay only for actual usage with automatic scaling.

This skill provides guidance for quality focused software development and architecture. It is based on Clean Architecture and Domain Driven Design principles.

Route through domain routers before using individual tools. Routers abstract tool selection.

Implement proven backend architecture patterns for maintainable, testable, and scalable systems. \n \n Covers three core patterns: Clean Architecture (layered dependency inward), Hexagonal Architecture (ports and adapters), and Domain-Driven Design (bounded contexts, aggregates, value objects) \n Includes complete directory structures, code examples, and implementation patterns for Python backends using FastAPI, asyncpg, and similar frameworks \n Demonstrates practical separation of concerns: do

Decision framework and service comparison patterns for architecting across AWS, Azure, GCP, and OCI. \n \n Includes detailed service mapping tables across compute, storage, and database categories to identify equivalent offerings and best-of-breed selections \n Four core multi-cloud patterns: single provider with disaster recovery, best-of-breed service selection, geographic distribution, and cloud-agnostic abstraction layers \n Cloud-agnostic alternatives using Kubernetes, PostgreSQL, Apache Ka