skills.rest
zzafergok
zzafergokCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

rag-implementation

Implement retrieval-augmented generation pipelines with vector databases, embeddings, and reranking for LLM applications.

Updated Jan 3, 2026
Search Tags:#semantic search#embeddings#langchain#llm#rag#vector database#retrieval
One-click install
npx skills add https://github.com/zzafergok/soonur --skill rag-implementation-zzafergok
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/zzafergok/soonur/tree/main/.agent/skills/rag-implementation
Command: npx skills add https://github.com/zzafergok/soonur --skill rag-implementation-zzafergok

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires langchain, openai, chromadb, pinecone-client, weaviate-client, sentence-transformers.

What problem does it solve? LLMs hallucinate and lack access to proprietary or current knowledge. This Skill guides you through building RAG systems that ground LLM responses in your own documents, enabling accurate Q&A with source citations. ## Core Features & Use Cases - Vector Database Integration: Configure Pinecone, Weaviate, Chroma, Qdrant, or FAISS to store and query document embeddings. - Retrieval Strategies: Implement dense, sparse (BM25), hybrid, multi-query, HyDE, and contextual compression retrieval patterns. - Chunking & Reranking: Apply recursive, token-based, semantic, and Markdown-aware splitting, plus cross-encoder reranking and MMR for diversity. - Use Case: Build a documentation assistant that answers questions about your company's internal docs, returning answers with citations to the exact source pages. ## Quick Start Ask the AI to build a RAG pipeline that loads documents from a folder, chunks them, indexes them in Chroma, and answers questions with source citations.

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 RAG system with LangChain?▼

Load documents with DirectoryLoader, split them with RecursiveCharacterTextSplitter, embed them with OpenAIEmbeddings, and store them in a vectorstore like Chroma. Then create a RetrievalQA chain combining an LLM with the vectorstore retriever to answer queries with source documents.

What vector database should I use for RAG?▼

Pinecone offers managed scalability, Chroma is lightweight for local development, Weaviate supports hybrid search, and FAISS works for local deployment. Choose based on hosting needs, query volume, and whether you need metadata filtering.

What chunk size works best for RAG document splitting?▼

Use 500-1000 tokens per chunk with 10-20% overlap to preserve context at boundaries. Smaller chunks improve retrieval specificity while larger chunks give the LLM more context; the Parent Document Retriever pattern combines both.

How do I reduce hallucinations in RAG applications?▼

Ground responses with prompts that instruct the LLM to answer only from provided context and say when information is insufficient. Add reranking with cross-encoders, return source citations, and include a verification step to check answer groundedness.

Why does my RAG retrieval return irrelevant results?▼

Irrelevant results usually stem from poor chunking, weak embedding models, or ambiguous queries. Fix this with hybrid search combining BM25 and dense retrieval, metadata filtering, multi-query expansion, and cross-encoder reranking of top candidates.

Related Skills

Explore More Software Engineering Skills→
iterationlayeriterationlayer
2

convert-document-for-rag-ingestion

Convert PDFs and web pages into clean Markdown for RAG ingestion.

Official
Intermediate
latestaiagentslatestaiagents
5

production-rag-checklist

Provide a production deployment checklist for RAG systems covering pipelines, monitoring, and security.

Official
Advanced
Giskard-AIGiskard-AI
5

rag-evaluator

Evaluate RAG systems for groundedness, relevance, and citation accuracy.

Official
Advanced
ricablericable

simpo-training

Train language models with SimPO using preference pairs without a reference model.

Community
Advanced