hybrid-search-implementation

Combines vector similarity and keyword retrieval with Reciprocal Rank Fusion for hybrid search systems.

Updated Feb 13, 2026
One-click install
npx skills add https://github.com/simplysmartai/5cypressautomation --skill hybrid-search-implementation-simplysmartai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hybrid-search-implementation
Source: https://github.com/simplysmartai/5cypressautomation/tree/main/agents/plugins/llm-application-dev/skills/hybrid-search-implementation
Command: npx skills add https://github.com/simplysmartai/5cypressautomation --skill hybrid-search-implementation-simplysmartai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires asyncpg, numpy, elasticsearch, sentence-transformers, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the limitations of using either vector search or keyword search alone, providing a more robust and accurate retrieval system by combining their strengths.

Core Features & Use Cases

  • Improved Recall: Combines semantic understanding with exact term matching to ensure more relevant results.
  • Flexible Fusion: Supports various methods like Reciprocal Rank Fusion (RRF) and linear combination for merging search results.
  • Use Case: Implementing a RAG system where semantic search might miss specific product codes, but keyword search can find them, leading to more comprehensive answers.

Quick Start

Use the hybrid-search-implementation skill to perform a search query for 'AI ethics in healthcare' using both vector and keyword methods.

Frequently Asked Questions about hybrid-search-implementation

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

FAQPage Schema
How do I combine vector search and keyword search for RAG systems?

To combine vector search and keyword search for RAG systems, you implement hybrid search strategies. This approach merges semantic vector similarity with exact keyword matching to improve recall and handle domain-specific vocabulary effectively.

What is Reciprocal Rank Fusion and how does it work for hybrid search?

Reciprocal Rank Fusion (RRF) is a flexible fusion method for hybrid search that merges result rankings. It combines the rankings from vector similarity and keyword-based retrieval methods to provide more accurate and robust search results.

How do I implement cross-encoder reranking in an information retrieval pipeline?

You can implement cross-encoder reranking in an information retrieval pipeline by utilizing advanced hybrid search techniques. This enhances retrieval accuracy by reordering the combined results from initial vector and keyword search methods.

Does Elasticsearch work with sentence-transformers for hybrid search?

Yes, Elasticsearch works with sentence-transformers for hybrid search. The implementation uses Elasticsearch for keyword-based retrieval and sentence-transformers to generate vector embeddings, combining them using methods like linear combination or RRF.

Why does vector search miss specific product codes in my search engine?

Vector search misses specific product codes because it focuses on semantic understanding rather than exact term matching. Adding a keyword-based retrieval component creates a hybrid search system that ensures exact terms are found.