rag-systems

Build RAG systems - embeddings, vector stores, chunking, and retrieval optimization

⭐5stars🔀1forks📁View on GitHub🕐Updated Jan 11, 2026⬇️2downloads

When & Why to Use This Skill

This Claude skill facilitates the end-to-end development of Retrieval-Augmented Generation (RAG) systems, enabling AI agents to deliver grounded and contextually accurate responses. It provides comprehensive tools for document chunking, embedding generation, and vector database management to optimize information retrieval from custom datasets and minimize LLM hallucinations.

Use Cases

  • Building specialized Q&A systems that provide answers based on private technical documentation, legal archives, or internal company wikis.
  • Implementing semantic search engines that understand user intent and context to find relevant information across large-scale unstructured datasets.
  • Optimizing AI response accuracy by grounding generation in specific, verified knowledge bases using advanced retrieval techniques like hybrid search and reranking.
  • Setting up and managing high-performance vector databases (e.g., Pinecone, Chroma, Weaviate) for efficient high-dimensional data storage and retrieval.
namerag-systems
descriptionBuild RAG systems - embeddings, vector stores, chunking, and retrieval optimization
sasmp_version"1.3.0"
bonded_agent03-rag-systems
bond_typePRIMARY_BOND
version"2.0.0"

RAG Systems

Build Retrieval-Augmented Generation systems for grounded responses.

When to Use This Skill

Invoke this skill when:

  • Building Q&A over custom documents
  • Implementing semantic search
  • Setting up vector databases
  • Optimizing retrieval quality

Parameter Schema

Parameter Type Required Description Default
task string Yes RAG goal -
vector_db enum No pinecone, weaviate, chroma, pgvector chroma
embedding_model string No Embedding model text-embedding-3-small
chunk_size int No Chunk size in chars 1000

Quick Start

from langchain_openai import OpenAIEmbeddings
from langchain_chroma import Chroma
from langchain_text_splitters import RecursiveCharacterTextSplitter

# 1. Split documents
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
chunks = splitter.split_documents(documents)

# 2. Create vector store
embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
vectorstore = Chroma.from_documents(chunks, embeddings)

# 3. Retrieve
docs = vectorstore.similarity_search("query", k=5)

Chunking Strategy

Content Type Size Overlap Rationale
Technical docs 500-800 100 Preserve code
Legal docs 1000-1500 200 Keep clauses
Q&A/FAQ 200-400 50 Atomic answers

Embedding Costs

Model Cost/1M tokens
text-embedding-3-small $0.02
text-embedding-3-large $0.13
Cohere embed-v3 $0.10

Troubleshooting

Issue Solution
Irrelevant results Improve chunking, add reranking
Missing context Increase k, use parent retriever
Hallucinations Add "only use context" prompt
Slow retrieval Add caching, reduce k

Best Practices

  • Always include source attribution
  • Use hybrid search (dense + BM25)
  • Implement reranking for quality
  • Evaluate with RAGAS metrics
  • llm-integration - LLM for generation
  • agent-memory - Memory retrieval
  • ai-agent-basics - Agentic RAG

References