skills.rest
vorynkavitaliy
vorynkavitaliyCOMMUNITY
Explore all skills by this author →
GitHub

hybrid-search-implementation

Combine vector similarity and keyword search with RRF, linear, or cascade fusion.

Updated Aug 23, 2026
Search Tags:#retrieval#vector search#hybrid search#rerank#keyword search#fusion
One-click install
npx skills add https://github.com/vorynkavitaliy/openclaw-assistent --skill hybrid-search-implementation-vorynkavitaliy
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hybrid-search-implementation
Source: https://github.com/vorynkavitaliy/openclaw-assistent/tree/main/.agents/skills/hybrid-search-implementation
Command: npx skills add https://github.com/vorynkavitaliy/openclaw-assistent --skill hybrid-search-implementation-vorynkavitaliy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hybrid search addresses the gap between vector-only and keyword-only retrieval by fusing both signals to improve recall without sacrificing precision.

Core Features & Use Cases

  • Fusion strategies: RRF, linear, cascade, and cross-encoder reranking to balance recall and precision.
  • Parallel hybrid retrieval: simultaneous vector and keyword querying with fusion to produce ranked results.
  • Real-world scenarios: building RAG pipelines, search engines for domain-specific vocabularies, and materials search with exact terms.

Quick Start

Provide a query and receive a fused ranked list from vector and keyword search pipelines.

Frequently Asked Questions about hybrid-search-implementation

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

FAQPage Schema
How does hybrid search improve retrieval performance for RAG systems?▼

Hybrid search improves retrieval performance by combining vector similarity and keyword search to increase recall without sacrificing precision. It fuses both signals to address the gap where either vector-only or keyword-only retrieval falls short.

What are the available fusion strategies for combining vector and keyword search?▼

Available fusion strategies include Reciprocal Rank Fusion (RRF), linear fusion, and cascade fusion. You can also apply optional cross-encoder reranking to boost precision after fusing vector and keyword search results.

When should I use hybrid search instead of vector-only or keyword-only retrieval?▼

Use hybrid search when neither vector-only nor keyword-only retrieval suffices, such as building RAG pipelines or search engines for domain-specific vocabularies. It handles materials search with exact terms where semantic similarity alone might miss critical keywords.

How do I implement parallel hybrid retrieval with fusion and reranking?▼

To implement parallel hybrid retrieval, run vector and keyword queries simultaneously and fuse the results into a ranked list. Apply cross-encoder reranking after fusion to further boost precision for your search pipeline.

Does hybrid search work for domain-specific vocabularies and exact term matching?▼

Yes, hybrid search works for domain-specific vocabularies by fusing vector similarity with keyword matching. This ensures exact terms are captured through keyword search while semantic context is preserved through vector search.

Related Skills

Explore More Software Engineering Skills→
winsorllcwinsorllc

Embedding Search

Provides semantic search across documents with Tunrsity and Tunrsity-like features.

Community
Intermediate
zilliztechzilliztech
3

rag-with-rerank

Add a cross-encoder reranking stage to RAG retrieval pipelines.

Official
Advanced
aisa-groupaisa-group
90

rag-implementation

Build retrieval-augmented generation systems with embeddings, vector stores, and reranking.

Official
Intermediate
emilneuraz-aiemilneuraz-ai

hybrid-search-implementation

Combines vector similarity and keyword matching for hybrid search.

Community
Advanced