Which agent frameworks does this skill support?

It covers multiple agentic patterns including Reasoning + Acting (ReAct), Function Calling patterns, Plan-and-Execute workflows, and specialized Multi-Agent team collaboration.

How does it help with RAG implementation?

It provides specific strategies for document ingestion, various chunking methods (semantic, recursive, fixed-size), hybrid search implementation, and contextual compression for better retrieval quality.

What architectural patterns are included in this skill?

The skill includes patterns for RAG (Retrieval-Augmented Generation), autonomous agent architectures like ReAct and Plan-and-Execute, multi-agent collaboration, and prompt management systems.

Does it provide guidance on selecting vector databases?

Yes, it includes a comparison and implementation guide for popular vector databases like Pinecone, Weaviate, ChromaDB, and pgvector based on scale and use-case requirements.

LLM Application Patterns

Name: LLM Application Patterns
Author: claudiodearaujo

byclaudiodearaujo

0•

데이터 과학 및 ML

Provides production-ready architectural patterns for building robust LLM applications, covering RAG pipelines, autonomous agents, and LLMOps observability.

This skill serves as a comprehensive blueprint for developing enterprise-grade AI applications by offering standardized implementation patterns for RAG (Retrieval-Augmented Generation), agentic workflows, and prompt management. It guides developers through critical design decisions such as semantic chunking strategies, vector database selection, and complex agent architectures like ReAct and Plan-and-Execute. By integrating industry best practices and LLMOps monitoring patterns, it enables the creation of scalable, reliable, and observable AI systems that go beyond simple chat interfaces to perform complex, multi-step reasoning tasks.

주요 기능

01Standardized prompt engineering templates with versioning and variable management

020 GitHub stars

03Advanced RAG pipeline architectures with hybrid search and reciprocal rank fusion

04Autonomous agent patterns including ReAct, Function Calling, and Multi-Agent orchestration

05Production-grade document ingestion and semantic chunking configurations

06Vector database integration patterns for Pinecone, Weaviate, and pgvector

사용 사례

01Building a high-accuracy RAG system for searching through complex technical documentation

02Developing autonomous AI agents capable of using external tools and mathematical reasoning

03Implementing a standardized LLM observation and prompt management layer for production apps

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/izacenter llm-app-patterns

For use in Claude.ai and ChatGPT

Download Skill