RAG Pipeline

Build production-ready Retrieval-Augmented Generation pipelines with document ingestion, chunking, and retrieval.

Category: AI & LLMs | Security status: verified | Rating: 4.9/5 | Version: 2.5.0

What RAG Pipeline does

RAG Pipeline provides an end-to-end framework for building Retrieval-Augmented Generation systems. Ingest documents (PDF, DOCX, HTML, Markdown), chunk them intelligently, generate embeddings, store in vector databases, and retrieve relevant context for LLM queries. Includes hybrid search (semantic + keyword), reranking, and citation tracking.

Use cases

Build Q&A systems over your company's knowledge base
Create AI assistants that cite their sources
Process and query large document collections
Build customer support bots grounded in your docs

Install RAG Pipeline

npx clawhub@latest install rag-pipeline

Frequently asked questions

What document formats does it support?

PDF, DOCX, XLSX, HTML, Markdown, plain text, and Notion exports. It handles tables, images with OCR, and code blocks.

How does hybrid search work?

It combines vector similarity search with BM25 keyword matching, then reranks results using a cross-encoder for maximum relevance.