Can this skill help with RAG evaluation?

Yes, it integrates the RAGAS framework to measure metrics like faithfulness, context relevance, and answer relevance to ensure output quality.

How does it improve retrieval precision?

It utilizes advanced techniques such as hybrid search (dense + sparse), reciprocal rank fusion, and cross-encoder reranking to ensure the most relevant context is retrieved.

What is the RAG Architect skill?

It is a specialized capability for Claude Code that helps engineers design, build, and optimize every stage of a Retrieval-Augmented Generation pipeline.

Does it support different vector databases?

Yes, it provides comparative analysis and implementation patterns for popular stores including Pinecone, Weaviate, Qdrant, Chroma, and pgvector.

RAG Architect

Name: RAG Architect
Author: borghei

byborghei

•

数据科学与机器学习

Designs, implements, and optimizes production-grade Retrieval-Augmented Generation (RAG) pipelines for high-performance AI applications.

RAG Architect is a specialized engineering skill designed to guide developers through the full lifecycle of building robust RAG systems. It provides expert guidance on corpus analysis, selecting optimal chunking strategies—from fixed-size to semantic—and choosing the right embedding models and vector databases like Pinecone or pgvector. By implementing advanced retrieval techniques such as hybrid search, cross-encoder reranking, and query transformations like HyDE, the skill ensures high precision and recall. It also emphasizes production readiness through RAGAS framework evaluation and the integration of critical guardrails for PII detection and hallucination prevention.

主要功能

01Comparative analysis for embedding models and vector database selection

02Automated evaluation using RAGAS metrics for faithfulness and relevance

0350 GitHub stars

04Advanced retrieval design with hybrid search and cross-encoder reranking

05Production-grade guardrails for PII, hallucinations, and source attribution

06Multi-strategy chunking optimization including semantic and recursive methods

使用场景

01Optimizing search precision for technical manuals using hybrid retrieval

02Building a secure internal knowledge base using enterprise documentation

03Evaluating and fixing hallucination issues in existing AI search pipelines

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add borghei/claude-skills rag-architect

For use in Claude.ai and ChatGPT

Download Skill