skills.rest
vasilyu1983
vasilyu1983COMMUNITY
Explore all skills by this author →
Website𝕏@vasilyuGitHub

ai-llm-rag-engineering

Design production-grade AI agents with MCP integration and agentic RAG.

73|16|Updated Nov 14, 2025
Search Tags:#vector database#LLM#RAG#hybrid search#chunking#reranking#retrieval-augmented generation
One-click install
npx skills add https://github.com/vasilyu1983/AI-Agents-public --skill ai-llm-rag-engineering
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-llm-rag-engineering
Source: https://github.com/vasilyu1983/AI-Agents-public/tree/main/frameworks/claude-code-kit/initial-setup/skills/ai-llm-rag-engineering
Command: npx skills add https://github.com/vasilyu1983/AI-Agents-public --skill ai-llm-rag-engineering

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

Static RAG systems often suffer from irrelevant retrieval, hallucinations, and poor context grounding. This Skill provides practical, production-grade RAG design patterns with modern advances for dynamic and intelligent retrieval.

Core Features & Use Cases

  • Advanced Retrieval: Implement hybrid retrieval (BM25 + vector) with cross-encoder reranking and Reciprocal Rank Fusion (RRF) for significant relevance gains.
  • Optimal Chunking: Utilize page-level or semantic chunking strategies for highest accuracy and recall, tailored to document structure.
  • Grounded Generation: Design contextual retrieval, context compression, and citation patterns to minimize hallucinations and ensure factual accuracy in LLM responses.

Quick Start

Use the ai-llm-rag-engineering skill to design a chunking strategy for a PDF document knowledge base, aiming for high accuracy and recall.

Frequently Asked Questions about ai-llm-rag-engineering

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reduce hallucinations in LLM responses using retrieval-augmented generation?▼

RAG grounds LLM outputs in retrieved documents, minimizing hallucinations through context compression and citation patterns. This Skill teaches design strategies for contextual retrieval and grounded generation to ensure factual accuracy in production systems.

What's the best way to implement hybrid search combining vector and BM25 retrieval?▼

Hybrid search merges vector similarity with keyword matching, then reranks results using cross-encoders and Reciprocal Rank Fusion for higher relevance. This Skill covers hybrid retrieval implementation with reranking patterns for significant accuracy gains.

How do I chunk documents optimally for RAG systems?▼

Optimal chunking uses page-level or semantic strategies tailored to document structure, maximizing retrieval accuracy and recall. This Skill provides chunking design patterns and reasoning for knowledge base RAG systems.

Can I build production-grade RAG agents with tool integration and multi-layer guardrails?▼

Yes. This Skill enables design of agentic RAG systems with MCP integration, tool definitions, action loops, memory systems, and security controls. It covers deployment templates, CI/CD workflows, and observability standards for tool-using agents.

What framework should I use to build multi-agent RAG systems?▼

LangGraph and MCP are core frameworks for orchestrating multi-agent systems with enforced tool integration and context management. This Skill teaches agent handoffs, evaluation harnesses, and deployment patterns across agentic architectures.

How do I know if my RAG retrieval is actually improving relevance?▼

Evaluation harnesses and observability plans measure retrieval quality and grounding accuracy. This Skill covers building evaluation frameworks and observability infrastructure to validate RAG system performance in production.

Related Skills

Explore More Software Engineering Skills→
devCharuzudevCharuzu

rag-architect

Design and deploy production-grade RAG pipelines with chunking, embeddings, vector stores, and evaluation.

Community
Advanced
vasilyu1983vasilyu1983
73

quality-debugging-troubleshooting

Diagnose software bugs and production incidents with systematic debugging workflows.

Community
Intermediate
vasilyu1983vasilyu1983
73

quality-observability-performance

Implement OpenTelemetry observability and SLOs for production AI agent systems.

Community
Advanced
48Nauts-Operator48Nauts-Operator
3

hybrid-search-implementation

Implement hybrid search combining vector similarity with BM25 ranking.

Community
Advanced