skills.rest
Axeloooo
AxelooooCOMMUNITY
Explore all skills by this author →
GitHub

rag-implementation

Implement RAG pipelines with vector retrieval and context-grounded generation.

Updated Oct 16, 2025
Search Tags:#semantic search#embeddings#rag#llm evaluation#vector database#langgraph#retrieval reranking
One-click install
npx skills add https://github.com/Axeloooo/TheDataBay --skill rag-implementation-axeloooo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/Axeloooo/TheDataBay/tree/main/.agents/skills/rag-implementation
Command: npx skills add https://github.com/Axeloooo/TheDataBay --skill rag-implementation-axeloooo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you reduce hallucinations and improve answer accuracy by grounding LLM outputs in relevant external documents retrieved at query time.

Core Features & Use Cases

  • Vector database + embeddings setup: Store document embeddings and retrieve them via semantic similarity using options like Pinecone, Weaviate, Milvus, Chroma, Qdrant, or pgvector.
  • Retrieval strategies: Improve recall and relevance with dense retrieval, sparse retrieval, hybrid search, multi-query retrieval, and HyDE.
  • Quality upgrades for retrieval: Use reranking (cross-encoders, Cohere rerank, MMR) and contextual compression to return better supporting context.
  • RAG orchestration and evaluation guidance: Build RAG pipelines (including LangGraph-style flows) and measure retrieval precision/recall and faithfulness.
  • Chunking and configuration patterns: Choose practical chunking methods (recursive, token-based, semantic, and header-based) and tune vector store configurations for your workload.

Quick Start

Use the rag-implementation skill to implement an end-to-end RAG pipeline for question answering over your documents with semantic retrieval and context-grounded generation.

Frequently Asked Questions about rag-implementation

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

FAQPage Schema
How do I reduce LLM hallucinations using a RAG pipeline?▼

A RAG pipeline reduces LLM hallucinations by retrieving relevant external documents via semantic search and injecting them into the generation context. This grounds LLM outputs in factual data for accurate answers.

What is the best way to implement semantic search over proprietary documents?▼

Implement semantic search by storing document embeddings in a vector database like Pinecone, Weaviate, Milvus, Chroma, Qdrant, or pgvector, then retrieving relevant context via dense, sparse, or hybrid similarity search.

Does the RAG pipeline support multi-query retrieval and HyDE strategies?▼

Yes, the RAG pipeline supports multi-query retrieval and HyDE alongside dense, sparse, and hybrid search. These retrieval strategies improve recall and relevance when fetching supporting context from the vector store.

How do I improve retrieval relevance with reranking and contextual compression?▼

Improve retrieval relevance by applying reranking techniques like cross-encoders, Cohere rerank, or MMR, alongside contextual compression. These methods filter and reorder retrieved documents to return better supporting context.

Can I use LangGraph to build RAG orchestration flows?▼

Yes, you can use LangGraph-style flows to build RAG orchestration pipelines. This approach manages the end-to-end flow from document retrieval to context-grounded generation and evaluation-driven testing.

What chunking methods should I use for document embeddings in a vector store?▼

Use recursive, token-based, semantic, or header-based chunking methods to split documents. Tuning these chunking patterns and vector store configurations optimizes retrieval performance for your specific workload.

Related Skills

Explore More Software Engineering Skills→
rizaldiemrizaldiem

chunking-strategy

Break large documents into semantically meaningful segments for retrieval and embedding.

Community
Advanced
DTMC-marketplaceDTMC-marketplace

ragas

Evaluate RAG pipelines on faithfulness, answer relevancy, context precision, and recall.

Official
Intermediate
hyukudanhyukudan
1

vector-databases

Guide vector database selection and usage for similarity search and RAG applications.

Community
Intermediate
h4vzzh4vzz
33

context-ranking

Rank retrieved context chunks by relevance, diversity, and utility.

Community
Advanced