Build RAG pipelines with vector databases, embeddings, and LLM generation.
npx skills add https://github.com/jg-chalk-io/Nora-LiveKit --skill moai-ml-rag
Please help me install this Agent Skill. Skill: moai-ml-rag Source: https://github.com/jg-chalk-io/Nora-LiveKit/tree/main/.claude/skills/moai-ml-rag Command: npx skills add https://github.com/jg-chalk-io/Nora-LiveKit --skill moai-ml-rag
This Skill provides guidance for building Retrieval-Augmented Generation systems with vector databases, embeddings, and scalable pipelines.
Set up a minimal LangChain-based RAG flow to retrieve docs and answer questions from a dataset.
High-intent search queries and answers about installing and using this skill.
Production RAG systems integrate semantic search, vector databases, and LLM generation to retrieve and ground answers. This Skill covers the full pipeline: offline indexing, chunking, embedding creation, vector-store management, retrieval ranking, and generation with context—applied to enterprise knowledge bases, customer support, and documentation retrieval.
Semantic search uses embeddings to find documents by meaning rather than exact word matches, enabling RAG systems to retrieve contextually relevant passages even when phrasing differs. This approach powers grounded question-answering workflows where up-to-date, relevant context matters more than keyword overlap.
Vector database selection depends on scale, latency, and deployment constraints. Pinecone offers managed hosting, Weaviate provides open-source flexibility, and Milvus scales to large datasets. This Skill guides embedding strategies and database considerations for enterprise RAG pipelines.
Yes. LangChain provides framework abstractions for RAG workflows, including retrieval chains and prompt construction with context. This Skill includes a quick-start LangChain-based flow to retrieve documents and generate grounded answers from your dataset.
Embedding dimension is a trade-off between retrieval quality and computational cost. This Skill covers model selection and dimension considerations for production RAG systems, helping you optimize for your scale and accuracy requirements.
RAG systems require offline indexing and periodic re-embedding to reflect updated documents. This Skill addresses vector-store management, chunking strategies, and retrieval-ranking logic to maintain grounded, up-to-date answers in enterprise workflows.
ai-kitchen-incChunk documents, generate embeddings, and index vectors in Pinecone for RAG.
Giskard-AIEvaluate RAG systems for groundedness, relevance, and citation accuracy.
rukikotooQuery the econometrics RAG knowledge base to retrieve relevant document chunks.
Mercurium-AnalyticsFuse BM25 lexical relevance with vector semantic similarity in pgsv.hybrid_search.
