RAG Workflow Planner

Design RAG pipeline architectures with chunking, embedding, retrieval, and reranking strategies.

9|1|Updated Mar 15, 2026

Search Tags:#architecture #embeddings #ai #llm #rag #retrieval #vector-store

Ready to run

npx skills add https://github.com/Notysoty/openagentskills --skill rag-workflow-planner

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: RAG Workflow Planner
Source: https://github.com/Notysoty/openagentskills/tree/main/skills/rag-workflow-planner
Command: npx skills add https://github.com/Notysoty/openagentskills --skill rag-workflow-planner

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Walks you through designing a complete, production-ready Retrieval-Augmented Generation (RAG) pipeline tailored to your use case, including chunking strategy, embedding model, vector store, retrieval approach, and reranking.

Core Features & Use Cases

Generates an architecture diagram, a component-by-component breakdown, and a starter code outline to accelerate build-out.
Helps plan and validate RAG pipelines for research, enterprise chatbots, and knowledge-base search, with step-by-step guidance and reuse-ready templates.
Useful for starting a new RAG project or redesigning an existing pipeline to improve retrieval quality and latency.

Quick Start

Describe your RAG use case and requirements to generate a complete, production-grade architecture.

Frequently Asked Questions about RAG Workflow Planner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I design a production-ready RAG pipeline for an enterprise chatbot?▼

To design a production-ready RAG pipeline, specify your chunking, embedding, vector store, retrieval, and reranking strategies tailored to your use case. This generates an architecture diagram, component breakdown, and starter code outline to accelerate build-out.

What is the best way to structure a retrieval-augmented generation architecture for knowledge-base search?▼

A strong retrieval-augmented generation architecture structures components by selecting appropriate chunking methods, embedding models, and vector stores. It delivers an ASCII architecture diagram and component-by-component breakdown to validate your knowledge-base search pipeline design.

How does chunking strategy affect RAG pipeline retrieval quality and latency?▼

Chunking strategy directly impacts RAG retrieval quality and latency by determining how source documents are segmented before embedding. Redesigning your chunking approach provides step-by-step guidance to optimize information density and improve overall pipeline performance.

Can I use this RAG planner to redesign an existing retrieval pipeline?▼

Yes, you can use this RAG planner to redesign an existing retrieval pipeline. By describing your current use case and latency issues, it validates and outlines improved chunking, vector store, and reranking strategies with justifications for each architectural choice.

Do I need a reranking strategy for my LLM retrieval pipeline?▼

A reranking strategy is essential for an LLM retrieval pipeline to refine initial vector store results and improve relevance. The planner outlines specific reranking approaches with justifications across common AI use cases like research and enterprise chatbots.

What components are included in a complete RAG architecture breakdown?▼

A complete RAG architecture breakdown includes chunking logic, embedding models, vector store configurations, retrieval mechanisms, and reranking layers. It provides a starter code outline alongside an ASCII diagram to visualize how each component interacts.