skills.rest
duanbiao2000
duanbiao2000COMMUNITY
Explore all skills by this author →
WebsiteGitHub

hybrid-search-implementation

Combine vector similarity with keyword search using RRF or linear fusion.

3|1|Updated Feb 3, 2026
Search Tags:#rag#search#retrieval#vector#hybrid#keyword#fusion
One-click install
npx skills add https://github.com/duanbiao2000/obsidianDoc26 --skill hybrid-search-implementation-duanbiao2000
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hybrid-search-implementation
Source: https://github.com/duanbiao2000/obsidianDoc26/tree/main/agents-main/plugins/llm-application-dev/skills/hybrid-search-implementation
Command: npx skills add https://github.com/duanbiao2000/obsidianDoc26 --skill hybrid-search-implementation-duanbiao2000

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hybrid search addresses the limitation of using only vector or only keyword search by enabling a combined retrieval approach that improves recall and relevance.

Core Features & Use Cases

  • Hybrid architecture patterns that fuse vector similarity with keyword matching
  • Use cases include building RAG systems, domain-specific search, and scenarios with terms like names or codes
  • Templates and example implementations for RRF, linear fusion, and re-ranking

Quick Start

Configure a hybrid search pipeline that pairs vector and keyword results, apply a fusion method of choice, and return the top results for a given query.

Frequently Asked Questions about hybrid-search-implementation

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

FAQPage Schema
What is hybrid search and how does it improve RAG retrieval?▼

Hybrid search combines vector similarity with keyword matching to improve recall and relevance, resolving the incomplete retrieval problem inherent in using either method alone within RAG systems.

How do I combine vector and keyword search results in a RAG pipeline?▼

You can combine vector and keyword search results by applying fusion methods like Reciprocal Rank Fusion (RRF) or Linear fusion, optionally adding a re-ranking step to return the top matched documents.

When should I use hybrid search instead of vector search only?▼

Use hybrid search instead of vector search when your queries involve specific terms like names or codes, where keyword matching provides necessary context that pure vector similarity cannot capture.

What is the difference between RRF and Linear fusion in hybrid search?▼

RRF (Reciprocal Rank Fusion) and Linear fusion are scoring methods used to merge ranked results from vector and keyword retrieval, with templates provided to configure either fusion approach based on relevance needs.

Does hybrid search support re-ranking after fusing vector and keyword results?▼

Yes, the hybrid search pipeline supports optional re-ranking after applying fusion methods, allowing you to further refine and order the combined vector and keyword search results for better accuracy.

Related Skills

Explore More Software Engineering Skills→
AI-Foundry-CoreAI-Foundry-Core
4

hybrid-search-implementation

Implements dual-branch hybrid search with RRF or linear fusion and optional cross-encoder reranking for retrieval workflows.

Official
Advanced
JeffallanJeffallan
10.9k

rag-architect

Design and implement retrieval-augmented generation pipelines with vector stores and hybrid search.

Community
Advanced
chicanoandres702chicanoandres702

hybrid-search-implementation

Fuse vector and keyword search results using RRF, Linear, Cross-encoder, or Cascade fusion.

Community
Advanced
mwathibenmwathiben

hybrid-search-implementation

Design hybrid retrieval fusing vector similarity with keyword matching.

Community
Advanced