skills.rest
Shubh2310-developer
Shubh2310-developerCOMMUNITY
Explore all skills by this author →
GitHub

rag-implementation

Implements retrieval-augmented generation pipelines with chunking, embeddings, vector stores, and reranking.

2|Updated Jan 10, 2026
Search Tags:#semantic search#embeddings#rag#vector search#chunking#reranking#retrieval augmented generation
One-click install
npx skills add https://github.com/Shubh2310-developer/ENGUNITYCORE --skill rag-implementation-shubh2310-developer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/Shubh2310-developer/ENGUNITYCORE/tree/main/.claude/skills/rag-implementation
Command: npx skills add https://github.com/Shubh2310-developer/ENGUNITYCORE --skill rag-implementation-shubh2310-developer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Naive chunk-and-embed RAG setups produce poor retrieval quality, inconsistent embeddings, and high latency. This Skill provides proven patterns for chunking, embedding, hybrid search, and reranking so retrieved context actually improves LLM answers. ## Core Features & Use Cases - Semantic Chunking: Chunk documents by meaning rather than fixed size, with overlap to preserve context across boundaries. - Hybrid Search: Combine dense vector search with sparse keyword search to improve recall across query types. - Contextual Reranking: Rerank retrieved documents with an LLM to surface the most relevant context. - Use Case: When building a document Q&A system over a large knowledge base, apply these patterns to avoid stale embeddings, mismatched embedding models, and single-strategy retrieval failures. ## Quick Start Ask the AI to design a RAG pipeline for your document collection with semantic chunking, hybrid search, and reranking.

Frequently Asked Questions about rag-implementation

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

FAQPage Schema
How do I implement RAG with vector search?▼

RAG implementation starts with chunking documents semantically, embedding them with a consistent model, and storing vectors in a vector store. At query time, retrieve candidates with hybrid search and rerank them with an LLM before passing context to the generator.

What chunking strategy works best for RAG?▼

Semantic chunking by meaning outperforms fixed-size chunking, which breaks context arbitrarily. Use recursive character splitting with overlap so information spanning boundaries is not lost during retrieval.

Why is my RAG retrieval quality poor?▼

Poor retrieval usually comes from bad chunking, mismatched embedding models between queries and documents, or relying on a single retrieval strategy. Combine dense vector search with sparse keyword search and add reranking to improve relevance.

How do I reduce RAG latency in responses?▼

RAG adds latency through embedding, retrieval, and reranking stages. Optimize by caching embeddings, limiting retrieved document counts, and applying reranking only to a small candidate set.

When should I not use RAG?▼

RAG is unnecessary overhead when the required knowledge fits in the model's context window or is already in its training data. It adds latency and complexity, so use it only when external or frequently updated documents must ground responses.

Related Skills

Explore More Software Engineering Skills→
vecearvecear

embedding-pipeline-builder

Build document embedding and retrieval pipelines for scalable semantic search.

Community
Advanced
Miosa-osaMiosa-osa
226

prompt-cache-optimizer

Cache and compress prompts to reduce token costs in LLM interactions.

Official
Advanced
boazcstrikeboazcstrike
3

context-optimization

Optimizes AI context windows via compaction, masking, KV-cache efficiency, and partitioning.

Community
Advanced
JeffallanJeffallan
10.9k

fine-tuning-expert

Automate fine-tuning workflows for large language models using LoRA/QLoRA and PEFT methods.

Community
Advanced