skills.rest
nimadorostkar
nimadorostkarCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

rag

Optimize retrieval-augmented generation pipelines with hybrid search, reranking, and semantic chunking.

25|3|Updated Jul 14, 2026
Search Tags:#embeddings#llm#rag#retrieval#grounding#vector-search
One-click install
npx skills add https://github.com/nimadorostkar/Claude-Skills-collection --skill rag-nimadorostkar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag
Source: https://github.com/nimadorostkar/Claude-Skills-collection/tree/main/skills/ai/rag
Command: npx skills add https://github.com/nimadorostkar/Claude-Skills-collection --skill rag-nimadorostkar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the common failure points in Retrieval-Augmented Generation, specifically distinguishing between retrieval errors and generation errors to ensure accurate, grounded AI responses.

Core Features & Use Cases

  • Component-wise Diagnosis: Systematically identify whether a wrong answer stems from poor retrieval or poor generation.
  • Pipeline Optimization: Implement hybrid search, reranking, and semantic chunking to improve retrieval recall.
  • Use Case: When your RAG system provides confident but incorrect answers, use this skill to evaluate your retrieval pipeline and implement grounded prompting to force the model to cite sources or admit ignorance.

Quick Start

Use the rag skill to evaluate my current retrieval pipeline and implement a grounded prompt that forces the model to cite its sources.

Frequently Asked Questions about rag

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

FAQPage Schema
How do I diagnose whether my RAG pipeline is failing due to retrieval errors or generation errors?▼

To improve RAG retrieval recall, this skill implements hybrid search, reranking, and semantic chunking strategies. These optimizations ensure your vector search pipeline captures more relevant context before passing documents to the large language model for generation.

What is the best way to force an LLM to cite sources and admit ignorance in grounded RAG?▼

The best way to ground RAG responses is by implementing grounded prompting that forces the model to cite sources or admit ignorance. This skill applies context grounding constraints to your generation pipeline, ensuring verifiable citations for high accuracy production AI applications.

Why does my retrieval-augmented generation system return confident but incorrect answers?▼

Yes, this skill diagnoses production AI applications requiring high accuracy and verifiable citations. It evaluates your document-based question answering systems against component-wise metrics to ensure pipeline reliability in production environments.

What are the limitations of relying solely on vector search for RAG document retrieval?▼

Vector search alone limits RAG document retrieval by missing exact keyword matches, reducing overall recall. This skill overcomes this constraint by implementing hybrid search and reranking strategies to capture a broader, more accurate context scope.

Related Skills

Explore More Software Engineering Skills→
bethanychamberlainbethanychamberlain
2

rag

Configure LibreChat's RAG pipeline for document ingestion and retrieval.

Community
Advanced
azeemdigitalazeemdigital

ai-seo

Audit brand visibility in AI-generated search answers across major answer engines.

Community
Advanced
zzafergokzzafergok
1

prompt-engineering-patterns

Design reusable prompt patterns for few-shot learning and chain-of-thought prompting.

Community
Advanced
neverinfamousneverinfamous
19

rag-pipelines

Design RAG pipelines with chunking, hybrid search, and cross-encoder reranking.

Community
Advanced