skills.rest
tankpkg
tankpkgOFFICIAL
Explore all skills by this author →
GitHub

@tank/rag-production

Design production RAG pipelines with chunking, retrieval, reranking, and context assembly.

1|1|Updated Feb 20, 2026
Search Tags:#embeddings#rag#vector database#evaluation#retrieval#chunking#reranking
One-click install
npx skills add https://github.com/tankpkg/packages --skill tank-rag-production
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: @tank/rag-production
Source: https://github.com/tankpkg/packages/tree/main/skills/rag-production
Command: npx skills add https://github.com/tankpkg/packages --skill tank-rag-production

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you design, debug, and operate production retrieval-augmented generation systems that return accurate answers from your own documents instead of relying on model memory alone.

Core Features & Use Cases

  • Chunking strategy selection: Choose between recursive, semantic, document-aware, and parent-child chunking for articles, code, legal text, FAQs, and structured docs.
  • Retrieval and ranking design: Apply vector search, hybrid search, metadata filtering, reranking, HyDE, multi-query retrieval, query decomposition, and contextual compression.
  • Production readiness: Plan embedding models, vector databases, context assembly, evaluation with RAGAS or DeepEval, caching, streaming, monitoring, and fallback paths.
  • Use case: A team building an internal knowledge assistant can use this Skill to move from a basic prototype to a measurable, cost-controlled, citation-ready RAG system.

Quick Start

Ask for a production RAG architecture for your corpus, and specify your document types, target vector database, latency budget, and evaluation goals.

Frequently Asked Questions about @tank/rag-production

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

FAQPage Schema
How do I build a production RAG pipeline that keeps retrieval accurate and cost-controlled?▼

To build a production RAG pipeline, you combine chunking, embeddings, retrieval, reranking, and context assembly while applying continuous quality measurement, metadata filtering, caching, and fallback strategies to keep retrieval accurate and cost-controlled.

What is the best chunking strategy for documents in a knowledge base?▼

The best chunking strategy depends on your document types: choose recursive, semantic, document-aware, or parent-child chunking to handle articles, code, legal text, FAQs, and structured docs effectively within your RAG system.

How does hybrid lexical-vector retrieval improve RAG search results?▼

Hybrid lexical-vector retrieval improves RAG search results by combining keyword-based matching with vector search, applying metadata filtering, reranking, and multi-query retrieval to ensure the most contextually relevant documents are assembled for generation.

Can I evaluate RAG pipeline quality using RAGAS or DeepEval?▼

Yes, you can evaluate RAG pipeline quality using RAGAS or DeepEval by integrating evaluation-driven optimization to continuously measure retrieval accuracy, faithfulness, and context relevance across your document corpus.

Does this approach support agentic and graph RAG architectures?▼

Yes, this RAG design approach supports agentic and graph RAG architectures, applying to document search, knowledge bases, multimodal ingestion, and hybrid retrieval with production operations that maintain accurate and faithful generation.

What fallback strategies do I need for production RAG operations?▼

Production RAG operations require fallback strategies including metadata filtering, caching, streaming, and monitoring paths to maintain retrieval accuracy and cost-control when primary vector database retrieval or embedding models underperform.

Related Skills

Explore More Software Engineering Skills→
andikarachmanandikarachman
14

data-preprocessing

Preprocess raw data into clean, analysis-ready datasets with pandas and numpy.

Community
Advanced
Veloxia-agencyVeloxia-agency

rag-architect

Design production RAG pipelines with chunking, embeddings, and vector databases.

Community
Advanced
ccfccf

embedding-strategies

Select and optimize embedding models for semantic search and RAG applications.

Community
Intermediate
eyadsibaieyadsibai
7

rag-frameworks

Compare RAG frameworks like LangChain and LlamaIndex for document QA.

Community
Intermediate