Which vector databases are supported?

This skill covers popular options including Chroma for local development, Pinecone for managed services, and Weaviate for hybrid search, alongside libraries like FAISS.

Can I combine keyword search with semantic search?

Yes, the skill includes patterns for Hybrid Search, which combines BM25 keyword matching with dense vector retrieval to improve the relevance of retrieved results.

How does this skill handle long documents?

It provides strategies for recursive character splitting, token-based splitting, and semantic chunking to ensure data fits optimally within the LLM's context window.

How do I reduce hallucinations in my LLM app?

By using retrieval strategies and grounded prompt engineering techniques included in this skill, the LLM is instructed to answer based strictly on the provided context.

What is RAG and why should I use it?

Retrieval-Augmented Generation (RAG) allows LLMs to access external data not in their training set, ensuring responses are factual, up-to-date, and grounded in your specific documents.

RAG Implementation Toolkit

Name: RAG Implementation Toolkit
Author: HermeticOrmus

byHermeticOrmus

0•

数据科学与机器学习

Builds advanced Retrieval-Augmented Generation (RAG) systems using vector databases and semantic search to ground AI responses in proprietary data.

This skill provides a comprehensive framework for implementing Retrieval-Augmented Generation (RAG) within LLM applications, enabling developers to connect AI models with external knowledge bases. It covers the entire pipeline from document loading and recursive chunking to advanced retrieval strategies like hybrid search, multi-query generation, and reranking. By integrating this skill, users can build reliable Q&A systems, documentation assistants, and research tools that provide accurate, source-cited responses while significantly reducing model hallucinations through grounded context.

主要功能

01Vector database integration including Chroma, Pinecone, and Weaviate

02Contextual compression and reranking for improved accuracy

03Advanced document chunking and metadata filtering strategies

04Grounded prompt engineering templates with source citations

05Hybrid and semantic search retrieval techniques

060 GitHub stars

使用场景

01Building Q&A systems over proprietary internal business documents

02Creating documentation assistants with source-backed answers

03Developing research tools that perform semantic searches across large datasets

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add hermeticormus/after-the-third-cup rag-implementation

For use in Claude.ai and ChatGPT

Download Skill