skills.rest
giosuetedeschi-spec
giosuetedeschi-specCOMMUNITY
Explore all skills by this author →
GitHub

rag-implementation

Implement semantic chunking, hybrid search, and reranking for RAG systems.

Updated Jan 12, 2026
Search Tags:#semantic search#embeddings#rag#vector search#hybrid search#reranking
One-click install
npx skills add https://github.com/giosuetedeschi-spec/bobu-website --skill rag-implementation-giosuetedeschi-spec
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-implementation
Source: https://github.com/giosuetedeschi-spec/bobu-website/tree/main/.claude/skills/rag-implementation
Command: npx skills add https://github.com/giosuetedeschi-spec/bobu-website --skill rag-implementation-giosuetedeschi-spec

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the common failures in naive RAG implementations, such as poor retrieval quality and irrelevant context, by providing advanced strategies for chunking, embedding, and reranking.

Core Features & Use Cases

  • Sophisticated Chunking: Implements semantic chunking to ensure context is preserved rather than split arbitrarily.
  • Hybrid Search: Combines dense vector embeddings with sparse keyword search to improve retrieval precision.
  • Contextual Reranking: Uses LLMs to rerank retrieved documents, ensuring only the most relevant information is passed to the final generation step.

Quick Start

Apply the rag-implementation skill to analyze my current document retrieval pipeline and suggest improvements for 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 improve retrieval accuracy in my RAG system?▼

Improve RAG retrieval accuracy by implementing semantic chunking to preserve context, combining dense vector embeddings with sparse keyword search for hybrid retrieval, and applying LLM-based contextual reranking to filter the most relevant documents before generation.

What is the best way to prevent irrelevant context in retrieval-augmented generation?▼

Prevent irrelevant context in retrieval-augmented generation by using contextual reranking, which leverages LLMs to evaluate and reorder retrieved documents, ensuring only high-fidelity information is passed to the final generation step.

How does hybrid search optimize document retrieval pipelines?▼

Hybrid search optimizes document retrieval by combining dense vector embeddings for semantic understanding with sparse keyword search for exact term matching, significantly improving overall retrieval precision and reducing irrelevant results.

Why does naive RAG produce poor retrieval quality?▼

Naive RAG produces poor retrieval quality due to arbitrary document splitting that destroys context and reliance on single-method search, which fails to capture complex semantic relationships and retrieve highly relevant information.

What chunking strategy preserves context for vector search?▼

Semantic chunking preserves context for vector search by splitting documents based on meaning and topical boundaries rather than arbitrary character counts, ensuring embeddings capture complete and coherent information.

How do I manage latency in large-scale RAG applications?▼

Manage latency in large-scale RAG applications by optimizing embedding consistency and applying rigorous retrieval optimization strategies to streamline data processing, ensuring high-fidelity context is delivered to LLMs efficiently.

Related Skills

Explore More Software Engineering Skills→
zainhaszainhas
2

together-embeddings

Generate text embeddings via the Together AI embeddings API.

Community
Intermediate
floflo777floflo777
33

rag-eval

Evaluate RAG systems using retrieval, generation, and latency metrics.

Community
Intermediate
Kingly-AgencyKingly-Agency

rag-implementation

Build RAG pipelines that ground LLM responses with retrieved external knowledge.

Official
Advanced
leonardoteodorooleonardoteodoroo

hybrid-search-implementation

Implement hybrid search combining vector similarity with keyword matching for RAG systems.

Community
Advanced