skills.rest
BagelHole
BagelHoleCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

rag-infrastructure

Build and operate RAG infrastructure with vector stores and hybrid search.

46|4|Updated Jan 27, 2026
Search Tags:#semantic search#information retrieval#llm#rag#vector database#embedding
One-click install
npx skills add https://github.com/BagelHole/DevOps-Security-Agent-Skills --skill rag-infrastructure
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-infrastructure
Source: https://github.com/BagelHole/DevOps-Security-Agent-Skills/tree/main/infrastructure/local-ai/rag-infrastructure
Command: npx skills add https://github.com/BagelHole/DevOps-Security-Agent-Skills --skill rag-infrastructure

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sentence-transformers, qdrant-client, langchain, cohere, openai, fastembed, redis, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the complexity of building and operating Retrieval-Augmented Generation (RAG) infrastructure, enabling users to create knowledge base Q&A systems and semantic search over large document collections.

Core Features & Use Cases

  • RAG Pipeline: Covers the end-to-end process from document ingestion to LLM response generation.
  • Embedding & Vector Stores: Integrates with various embedding models and vector databases (Qdrant, Weaviate, Pinecone, pgvector).
  • Hybrid Search: Implements combined dense and sparse (BM25) retrieval for improved accuracy.
  • Reranking: Enhances relevance by reranking retrieved documents before LLM consumption.
  • Use Case: Develop a Q&A system for internal company documentation that provides accurate, context-aware answers to employee queries.

Quick Start

Use the rag-infrastructure skill to ingest documents into a Qdrant vector store using the provided Python scripts.

Frequently Asked Questions about rag-infrastructure

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

FAQPage Schema
How do I build a RAG pipeline for internal documentation Q&A?▼

To build a RAG pipeline for documentation Q&A, you need to handle document ingestion, chunking, embedding, and retrieval. This skill provides the infrastructure to manage these end-to-end processes, enabling accurate, context-aware LLM responses over large document collections.

How does hybrid search improve retrieval accuracy in RAG systems?▼

Hybrid search improves retrieval accuracy by combining dense vector search with sparse BM25 retrieval. This approach captures both semantic meaning and exact keyword matches, ensuring that the RAG pipeline fetches the most relevant context before the LLM generates an answer.

Can I use Qdrant and Sentence Transformers for vector stores and embeddings?▼

Yes, you can use Qdrant as your vector store and Sentence Transformers for generating embeddings. The skill integrates with these dependencies to manage document ingestion and operate semantic search capabilities for grounded LLM responses.

What is the best way to rerank retrieved documents before LLM consumption?▼

The best way to rerank retrieved documents is to apply a dedicated reranking step after hybrid search. This enhances relevance by reordering the fetched chunks, ensuring the LLM consumes only the most pertinent information to generate accurate answers.

Does this RAG infrastructure support production deployment with OpenAI and Redis?▼

Yes, this RAG infrastructure supports production deployment patterns using OpenAI endpoints and Redis. It manages the complete pipeline from document ingestion to LLM response generation, integrating with these dependencies for scalable and grounded Q&A systems.

Related Skills

Explore More Software Engineering Skills→
patricio0312revpatricio0312rev
5

vector-db-setup

Configure Pinecone, Chroma, pgvector, and Qdrant vector databases for semantic search.

Community
Advanced
woody1234567woody1234567
1

langchain-rag

Construct and execute a RAG pipeline with document loading, splitting, embedding, and storage.

Community
Intermediate
Axel-DaMageAxel-DaMage
2

rag-architect

Design and implement RAG systems with document chunking, embeddings, and hybrid search.

Community
Advanced
DrLuggelsDrLuggels
3

similarity-search-patterns

Provide patterns for similarity search with vector databases like Pinecone and Qdrant.

Community
Advanced