embedding-strategy

Official

Asymmetric embeddings for smarter retrieval.

AuthorAgentient
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Documents and queries often suffer from semantic mismatch when the same embedding path is used for both ingestion and querying. This Skill introduces an asymmetric embedding strategy to separate ingestion embeddings from query embeddings, reducing retrieval errors and improving relevance.

Core Features & Use Cases

  • Asymmetric embedding paths: Ingestion uses RETRIEVAL_DOCUMENT while queries use RETRIEVAL_QUERY to maintain task-type separation.
  • Semantic mismatch prevention and improved retrieval quality across document collections.
  • Model selection and configuration for embedding pipelines, plus validation of task_type usage.

Quick Start

Configure the embedding pipeline to route documents through RETRIEVAL_DOCUMENT and queries through RETRIEVAL_QUERY.

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: embedding-strategy
Download link: https://github.com/Agentient/vibekit/archive/main.zip#embedding-strategy

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 471,000+ vetted skills library on demand.