What is the LLM Application Patterns skill?

It is a specialized capability for Claude Code that provides standardized, production-grade patterns for building AI applications, covering RAG, agent architectures, and LLM observability.

Does this skill help with vector database selection?

Yes, it includes comparison guides and implementation patterns for popular vector databases like Pinecone, Weaviate, ChromaDB, and pgvector.

Can I use this to build multi-agent systems?

Absolutely. The skill provides patterns for agent coordination, task decomposition, and feedback loops between specialized agents (e.g., researchers, analysts, and critics).

How does it improve RAG retrieval quality?

It offers advanced strategies such as hybrid search (semantic + keyword), multi-query retrieval, and contextual compression to ensure the LLM receives the most relevant data.

LLM Application Patterns

Name: LLM Application Patterns
Author: claudiodearaujo

byclaudiodearaujo

•

数据科学与机器学习

Provides production-ready architectural patterns and implementation guides for building advanced RAG pipelines and autonomous AI agents.

This skill serves as a comprehensive blueprint for developers building sophisticated LLM applications. It offers detailed implementation strategies for Retrieval-Augmented Generation (RAG), including chunking and hybrid search, alongside diverse agent architectures like ReAct and multi-agent collaboration. By providing standardized patterns for prompt management and LLMOps, it enables the creation of scalable, observable, and high-performing AI systems within the Claude Code environment, ensuring best practices are followed from ingestion to response generation.

主要功能

011 GitHub stars

02Comprehensive RAG pipeline blueprints with advanced retrieval strategies

03Multiple autonomous agent architectures including ReAct and Plan-and-Execute

04Multi-agent collaboration and coordination workflow patterns

05Detailed document chunking and embedding model selection guides

06Prompt IDE patterns for versioning, templating, and A/B testing

使用场景

01Designing a complex document retrieval system with hybrid search and RRF

02Building an autonomous AI agent capable of using specialized external tools

03Setting up a multi-agent team for collaborative task solving and review

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/sistema-de-narra-o-de-livro llm-app-patterns

For use in Claude.ai and ChatGPT

Download Skill