skills.rest
Jhabbig
JhabbigCOMMUNITY
Explore all skills by this author →
GitHub

rag-implementation

Build retrieval-augmented generation systems that ground LLM answers in external knowledge.

Updated Apr 5, 2026
Search Tags:#embeddings#rag#citations#vector search#langgraph#reranking#semantic retrieval
One-click install
npx skills add https://github.com/Jhabbig/Habbig --skill rag-implementation-jhabbig
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/Jhabbig/Habbig/tree/main/.claude/plugins/wshobson/llm-application-dev/skills/rag-implementation
Command: npx skills add https://github.com/Jhabbig/Habbig --skill rag-implementation-jhabbig

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you build retrieval-augmented generation systems that answer questions with grounded, source-backed knowledge instead of relying only on model memory.

Core Features & Use Cases

  • Knowledge-Grounded QA: Connect LLMs to proprietary documents, policies, manuals, or research collections for accurate answers.
  • Retrieval Pipelines: Implement vector search, hybrid retrieval, multi-query strategies, parent-document retrieval, and context compression.
  • Quality Improvements: Add reranking, metadata filtering, and citation-aware prompting to reduce hallucinations and improve relevance.
  • Use Case: A product team can turn a folder of technical docs into a support assistant that finds the right passages and answers with citations.

Quick Start

Use the rag-implementation skill to design a retrieval-augmented question-answering system over your documents with grounded answers and source citations.

Frequently Asked Questions about rag-implementation

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

FAQPage Schema
How do I build a retrieval-augmented generation system that grounds LLM answers in my documents?▼

To build a retrieval-augmented generation system, you need vector search to retrieve relevant passages, reranking to improve relevance, and prompt orchestration to deliver cited, context-aware responses from your external knowledge base.

What is the best way to add citations to LLM answers over proprietary documents?▼

Adding citations to LLM answers requires citation-aware prompting combined with semantic retrieval. This grounds responses in specific document passages, reducing hallucinations and providing source-backed knowledge for accurate question answering.

How does vector search and reranking improve semantic retrieval for chatbots?▼

Vector search finds relevant documents through embeddings, while reranking refines those results to improve semantic retrieval quality. Together they ensure chatbots retrieve the right passages to generate context-aware responses.

Can I use LangGraph for prompt orchestration in a document question answering system?▼

LangGraph can be used for prompt orchestration in a document question answering system. It helps coordinate retrieval pipelines, reranking, and context compression to deliver grounded answers with source citations.

What retrieval pipeline strategies work best for knowledge-grounded research assistants?▼

Effective retrieval pipelines for research assistants include hybrid retrieval, multi-query strategies, parent-document retrieval, and context compression. These strategies help semantic search systems find precise passages for grounded answers.

Why does my retrieval-augmented generation system return irrelevant results and hallucinate answers?▼

Retrieval-augmented generation systems hallucinate or return irrelevant results when lacking proper reranking and metadata filtering. Adding these quality improvements alongside citation-aware prompting reduces hallucinations and boosts relevance.

Related Skills

Explore More Software Engineering Skills→
GregsGreyCodeGregsGreyCode
6

chroma

Store embeddings locally with metadata filtering and similarity search.

Community
Advanced
antgroupantgroup
34

context-enhancer

Enhances user message intent and semantic clarity for AI response accuracy using Python 3.8+ templates.

Official
Basic
alexander-danilenkoalexander-danilenko
14

rag-architect

Design retrieval-augmented generation architectures with hybrid search and reranking.

Community
Advanced
GitnappGitnapp

unsloth

Optimize LLM fine-tuning workflows with Unsloth-based LoRA and QLoRA training.

Community
Advanced