multi-vector-search

Official

Search across multiple fields to boost relevance.

Authorzilliztech
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill enables searching across multiple semantic facets of each item, such as title, description, reviews, and images, to provide more accurate results than single-vector search.

Core Features & Use Cases

  • Separate field embeddings for each semantic aspect (title, description, reviews, image) to preserve individual meaning.
  • Flexible fusion strategies (RRF or weighted rankers) to tailor results based on user intent.
  • Use cases include product catalogs, document repositories, and Q&A systems where different parts of an item contribute differently to relevance.

Quick Start

Index your data with separate fields and embeddings (title, description, reviews, image). Then perform a multi-vector search by providing a query and selecting a mode (balanced, title, description, or visual). Example: "Find products matching 'smartphone with good camera' across title and reviews while including image similarity."

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: multi-vector-search
Download link: https://github.com/zilliztech/milvus-marketplace/archive/main.zip#multi-vector-search

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.