[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-rag-pipelines-en":3,"seo:pack:rag-pipelines:en":88},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":87},"rag-pipelines","📚","#3B82F6","stable","Stable","RAG Pipelines","Quivr, RAGFlow, GraphRAG, plus production best-practices. Skip the bad first-retrieval architecture.",[16,28,37,45,52,60,70,80],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},322,"96223597-08c2-4e60-b84e-0c4779641933","quivr-opinionated-rag-framework-any-llm-96223597","Quivr — Opinionated RAG Framework for Any LLM","Quivr is an opinionated RAG framework supporting any LLM, multiple file types, and customizable retrieval. 39.1K+ stars. Apache 2.0.","Script Depot",310,0,"en","script","Script",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":22,"view_count":34,"vote_count":24,"lang_type":25,"type":35,"type_label":36},245,"7785d7a8-fc57-42ab-ba6b-4a970404fadc","ragflow-deep-document-understanding-rag-engine-7785d7a8","RAGFlow — Deep Document Understanding RAG Engine","Open-source RAG engine with deep document understanding. Parses complex PDFs, tables, images. Agent-powered Q&A with citations. Multi-model. 77K+ stars.",381,"skill","Skill",{"id":38,"uuid":39,"slug":40,"title":41,"description":42,"author_name":43,"view_count":44,"vote_count":24,"lang_type":25,"type":35,"type_label":36},418,"ac77668d-1767-4b86-ac8c-1c050166d21b","graphrag-knowledge-graph-rag-microsoft-ac77668d","GraphRAG — Knowledge Graph RAG by Microsoft","Build knowledge graphs from documents for smarter RAG. Local and global search over entity relationships. By Microsoft Research. 31K+ stars.","Microsoft AI",374,{"id":46,"uuid":47,"slug":48,"title":49,"description":50,"author_name":22,"view_count":51,"vote_count":24,"lang_type":25,"type":35,"type_label":36},242,"b0f93b10-3339-4ca0-ad20-d6335a3d7785","kotaemon-open-source-rag-document-chat-b0f93b10","Kotaemon — Open-Source RAG Document Chat","Clean, open-source RAG tool for chatting with your documents. Supports PDF, DOCX, web pages. Multi-model, citation, and multi-user. Self-hostable. 25K+ stars.",333,{"id":53,"uuid":54,"slug":55,"title":56,"description":57,"author_name":58,"view_count":59,"vote_count":24,"lang_type":25,"type":35,"type_label":36},1306,"e0e719be-37db-11f1-9bc6-00163e2b0d79","verba-golden-ragtriever-weaviate-e0e719be","Verba — The Golden RAGtriever by Weaviate","Verba is an open-source RAG (Retrieval-Augmented Generation) chatbot from the Weaviate team. Drop in PDFs, web pages, or notes; pick a model (OpenAI, Ollama, Anthropic); and get a polished chat UI with semantic search built in.","AI Open Source",354,{"id":61,"uuid":62,"slug":63,"title":64,"description":65,"author_name":66,"view_count":67,"vote_count":24,"lang_type":25,"type":68,"type_label":69},654,"7ded33e8-464c-4c8f-b3de-6dcf14c0eaf4","rag-best-practices-production-pipeline-guide-2026-7ded33e8","RAG Best Practices — Production Pipeline Guide 2026","Comprehensive guide to building production RAG pipelines. Covers chunking strategies, embedding models, vector databases, retrieval techniques, evaluation, and common pitfalls with code examples.","Prompt Lab",332,"prompt","Prompt",{"id":71,"uuid":72,"slug":73,"title":74,"description":75,"author_name":76,"view_count":77,"vote_count":24,"lang_type":25,"type":78,"type_label":79},635,"f73611a0-142f-4364-97dc-b57eb03473ad","tavily-search-api-built-ai-agents-rag-f73611a0","Tavily — Search API Built for AI Agents & RAG","Search API designed specifically for AI agents and RAG pipelines. Returns clean, LLM-ready results with content extraction, no HTML parsing needed. Official MCP server available. 5,000+ stars.","Tavily",337,"mcp","MCP",{"id":81,"uuid":82,"slug":83,"title":84,"description":85,"author_name":22,"view_count":86,"vote_count":24,"lang_type":25,"type":35,"type_label":36},205,"761bd107-7156-4c62-b268-62a3fb9818dc","haystack-ai-orchestration-search-rag-761bd107","Haystack — AI Orchestration for Search & RAG","Open-source AI orchestration framework by deepset. Build production RAG pipelines, semantic search, and agent workflows with modular components. 25K+ GitHub stars.",263,"tokrepo install pack\u002Frag-pipelines",{"pageType":89,"pageKey":8,"locale":25,"title":90,"metaDescription":91,"h1":13,"tldr":92,"bodyMarkdown":93,"faq":94,"schema":110,"internalLinks":119,"citations":132,"wordCount":145,"generatedAt":146},"pack","RAG Pipelines: Quivr, RAGFlow, GraphRAG production patterns","Skip the bad first-retrieval architecture. Quivr, RAGFlow, GraphRAG plus chunking, reranking, and eval patterns that survive production. Install via TokRepo.","Eight retrieval-augmented generation assets — open-source engines (Quivr, RAGFlow, GraphRAG) plus the chunking, reranking, and evaluation patterns that separate a demo from a production RAG system.","## What's in this pack\n\nMost teams ship their first RAG demo in a weekend and then spend six months untangling why it gives subtly wrong answers. This pack collects the **eight assets** that get you past that wall: three production-grade engines, three retrieval\u002Findexing patterns, and two evaluation tools.\n\n| # | Asset | Layer | Why it's here |\n|---|---|---|---|\n| 1 | Quivr | full-stack RAG | the \"second brain\" reference implementation, MIT-licensed |\n| 2 | RAGFlow | full-stack RAG | deep document parsing — beats LangChain for tables\u002Fforms |\n| 3 | GraphRAG | retrieval | Microsoft's knowledge-graph approach for multi-hop questions |\n| 4 | Chunking patterns | indexing | semantic vs fixed-size vs recursive — when each wins |\n| 5 | Hybrid search | retrieval | BM25 + dense vectors, with reranking |\n| 6 | Cross-encoder reranker | retrieval | the single biggest precision lift you can drop in |\n| 7 | RAG eval harness | observability | golden-set + LLM-as-judge for nightly regression |\n| 8 | Citation enforcement | guardrails | refuse-to-answer when retrieval below threshold |\n\n## Why this matters\n\nVector search alone gets you ~70% of demo quality. The last 30% — the part users actually notice — comes from the *non-vector* layers: how you chunk, how you rerank, how you decide when retrieval failed and the LLM should refuse rather than hallucinate.\n\nThree failure modes show up in every RAG audit we run:\n\n1. **Chunking destroys context.** A naïve 512-token split breaks tables in half and orphans headings. RAGFlow's layout-aware parser solves this; pure-LangChain pipelines don't.\n2. **Top-k retrieval returns near-duplicates.** Cosine similarity loves to surface 5 paraphrases of the same paragraph. A cross-encoder rerank step (BGE-reranker, Cohere Rerank) cuts duplicate-payload by 60%+ on most corpora.\n3. **No multi-hop reasoning.** Single-shot vector lookup can't answer \"compare X across years 2022, 2023, and 2024.\" GraphRAG builds a knowledge graph at index time so traversal-based answers become possible.\n\n## Install in one command\n\n```bash\n# Install the entire pack\ntokrepo install pack\u002Frag-pipelines\n\n# Or pick the engine you want to start with\ntokrepo install quivr\ntokrepo install ragflow\ntokrepo install graphrag\n```\n\nThe TokRepo CLI normalizes setup files across the eight supported AI tools, so the engines come pre-configured to slot into your existing Claude Code, Cursor, or Codex CLI project.\n\n## Common pitfalls\n\n- **Treating RAG as \"embed everything.\"** The cheapest precision win is *not* indexing low-signal pages. Audit your corpus first; remove duplicates, navigation chrome, and outdated versions.\n- **Skipping the rerank step.** Adding a cross-encoder rerank on top-50 → top-5 typically lifts answer-correctness by 15-25 points on RAG benchmarks. Skipping it to \"save latency\" is almost always wrong.\n- **No eval harness.** If you can't run a golden-set regression, you can't tell whether your last prompt change made things better or worse. Build the eval before you scale the corpus.\n- **Storing chunks without parent context.** Always keep a pointer back to the source document and adjacent chunks; let the LLM expand if it needs more context.\n- **Picking a vector DB before knowing your scale.** Pinecone makes sense at 100M+ vectors; below 10M, Qdrant or Chroma on a single VM is faster, cheaper, and easier to debug.\n\n## When this pack alone isn't enough\n\nIf your bottleneck is *ingest quality* (PDFs, scans, multi-column layouts), pair this with the Document AI Pipeline pack — Surya\u002FDocling\u002FMinerU clean up the source before chunking. If your bottleneck is *evaluation*, layer the LLM Eval & Guardrails pack on top: DeepEval, Ragas, and Promptfoo plug into the eval harness here.\n\nFor storage: this pack is engine-agnostic — see the Vector DB Showdown pack to choose between Chroma, Weaviate, Pinecone, Qdrant, or txtai based on your latency, cost, and accuracy targets.",[95,98,101,104,107],{"q":96,"a":97},"Are these RAG engines free?","Quivr, RAGFlow, and GraphRAG are all open-source under permissive licenses (Apache 2.0 \u002F MIT). You self-host. The only paid components you might add are the embedding API (OpenAI, Cohere, Voyage) and a managed vector DB if you don't want to run your own. A laptop-scale demo costs nothing; a 10M-doc production deployment is dominated by the embedding bill, not the engine.",{"q":99,"a":100},"How does GraphRAG compare to vanilla RAG?","Vanilla RAG retrieves top-k chunks by vector similarity and stuffs them in the prompt — great for single-hop questions like \"what is X.\" GraphRAG builds an entity-relationship graph at index time, so it can answer multi-hop questions like \"how did X's role change across these documents.\" The trade-off: indexing is 5-10x more expensive and slower. Use GraphRAG when your queries are analytical, vanilla RAG when they're factual lookups.",{"q":102,"a":103},"Will this work with Cursor or Codex CLI?","Yes — these are server-side engines, not editor extensions. You run RAGFlow or Quivr as a service, then any AI coding tool that can call HTTP can query it. The TokRepo install drops the docker-compose and config files into your project so the same setup works across Claude Code, Cursor, Codex CLI, Cline, and the rest. The retrieval API is identical.",{"q":105,"a":106},"What's the difference between this pack and the Vector DB Showdown pack?","Vector DB Showdown answers \"where do my embeddings live\" — Chroma, Qdrant, Pinecone, Weaviate, etc. RAG Pipelines answers \"how do I retrieve and rerank from that storage to produce a correct answer.\" You pick one from each. Most production setups are Qdrant or pgvector underneath, with RAGFlow or a custom pipeline on top.",{"q":108,"a":109},"How do I know if my RAG is actually working?","Build a golden set of 50-200 question-answer pairs from real user queries. Run them nightly. Track three numbers: retrieval recall (did the right chunk appear in top-k), answer correctness (LLM-as-judge against the gold answer), and citation faithfulness (did the answer cite a real retrieved chunk). Without these three, you're flying blind. Pack 28 (LLM Eval & Guardrails) ships the harness.",{"@context":111,"@type":112,"name":13,"description":113,"numberOfItems":114,"publisher":115},"https:\u002F\u002Fschema.org","CollectionPage","Quivr, RAGFlow, GraphRAG and the production patterns that beat naive vector search.",8,{"@type":116,"name":117,"url":118},"Organization","TokRepo","https:\u002F\u002Ftokrepo.com",[120,124,128],{"url":121,"anchor":122,"reason":123},"\u002Fen\u002Fpacks\u002Fvector-db-showdown","Vector DB Showdown","the storage layer underneath any RAG pipeline",{"url":125,"anchor":126,"reason":127},"\u002Fen\u002Fpacks\u002Fdocument-ai-pipeline","Document AI Pipeline","the ingest layer that feeds RAG",{"url":129,"anchor":130,"reason":131},"\u002Fen\u002Fpacks\u002Fllm-eval-guardrails","LLM Eval & Guardrails","score retrieval quality before shipping",[133,137,141],{"claim":134,"source_name":135,"source_url":136},"GraphRAG combines knowledge graphs with retrieval for multi-hop reasoning","Microsoft GraphRAG","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fgraphrag",{"claim":138,"source_name":139,"source_url":140},"RAGFlow open-source engine for deep document understanding RAG","infiniflow\u002Fragflow","https:\u002F\u002Fgithub.com\u002Finfiniflow\u002Fragflow",{"claim":142,"source_name":143,"source_url":144},"Quivr personal generative AI second brain with RAG","QuivrHQ\u002Fquivr","https:\u002F\u002Fgithub.com\u002FQuivrHQ\u002Fquivr",630,"2026-05-02T15:00:00Z"]