rag-implementation

Develop RAG systems integrating vector databases, embedding models, and retrieval strategies.

1|Updated Feb 24, 2026

Search Tags:#semantic search #llm #rag #vector database #knowledge base #retrieval

Ready to run

npx skills add https://github.com/xurenlu/marstaff --skill rag-implementation-xurenlu

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/xurenlu/marstaff/tree/main/skills/rag-implementation
Command: npx skills add https://github.com/xurenlu/marstaff --skill rag-implementation-xurenlu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables the creation of advanced Retrieval-Augmented Generation (RAG) systems, allowing Large Language Models (LLMs) to access and utilize external knowledge bases for more accurate and grounded responses.

Core Features & Use Cases

Vector Databases: Integrates with various vector stores (Pinecone, Weaviate, Chroma, etc.) for efficient semantic search.
Embedding Models: Supports multiple embedding models for converting text to vectors.
Retrieval Strategies: Implements diverse retrieval methods like dense, sparse, hybrid, multi-query, and HyDE.
Reranking: Enhances retrieval quality using methods like cross-encoders and MMR.
Use Case: Build a Q&A system over your company's internal documentation, ensuring the LLM provides answers directly supported by the provided knowledge base.

Quick Start

Use the rag-implementation skill to build a Q&A system over proprietary documents.

Frequently Asked Questions about rag-implementation

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

FAQPage Schema

How do I build a knowledge-grounded LLM application using a vector database?▼

To build a knowledge-grounded LLM application, you integrate a vector database with embedding models and retrieval logic to supply external context. This enables semantic search over your proprietary documents for accurate LLM responses.

What retrieval strategies can I use for Retrieval-Augmented Generation systems?▼

Retrieval-Augmented Generation systems support diverse retrieval strategies including dense, sparse, hybrid, multi-query, and HyDE. You can further enhance retrieval quality using reranking methods like cross-encoders and MMR.

Can I use this RAG implementation with Pinecone, Weaviate, and Chroma vector stores?▼

Yes, this RAG implementation integrates with various vector stores including Pinecone, Weaviate, and Chroma. This compatibility allows you to perform efficient semantic search across your preferred vector database platform.

How do I create a document Q&A system over internal company documentation?▼

You can create a document Q&A system by applying embedding models to convert internal documentation into vectors and storing them in a vector database. The LLM then retrieves this external knowledge to answer queries directly from supported documents.

What's the best way to improve semantic search quality in LLM applications?▼

To improve semantic search quality in LLM applications, implement advanced reranking methods like cross-encoders and MMR alongside hybrid retrieval strategies. This ensures the LLM receives the most relevant context from your vector database.