skills.rest
arthrod
arthrodCOMMUNITY
Explore all skills by this author →
GitHub

sqlite-vec

Enable vector similarity search in SQLite with KNN queries and distance functions.

Updated Jul 18, 2026
Search Tags:#database#semantic search#embeddings#sqlite#vector search#knn
One-click install
npx skills add https://github.com/arthrod/conejo-skills --skill sqlite-vec-arthrod
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sqlite-vec
Source: https://github.com/arthrod/conejo-skills/tree/main/skills/sqlite-vec
Command: npx skills add https://github.com/arthrod/conejo-skills --skill sqlite-vec-arthrod

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sqlite3, numpy, and includes references (resource) components.

What problem does it solve?

This Skill solves the challenge of performing high-performance vector similarity searches directly within SQLite, eliminating the need for complex, external vector database infrastructure.

Core Features & Use Cases

  • Vector Storage: Native support for float32, int8, and bit vectors directly in virtual tables.
  • Similarity Search: Efficient KNN queries using L2, cosine, or hamming distance metrics.
  • Use Case: Build semantic search engines, recommendation systems, or multi-tenant RAG applications by embedding your data and querying it with standard SQL.

Quick Start

Use the sqlite-vec skill to create a virtual table named vec_items with a 768-dimensional embedding column and perform a nearest neighbor search.

Frequently Asked Questions about sqlite-vec

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

FAQPage Schema
How do I perform vector similarity search in SQLite?▼

Vector similarity search in SQLite is performed by loading the sqlite-vec extension to create virtual tables for float32, int8, or bit vectors, enabling KNN queries using L2, cosine, or hamming distance metrics.

Can I use SQLite for semantic search and RAG applications without an external vector database?▼

Yes, SQLite supports semantic search and multi-tenant RAG applications natively by storing embeddings in virtual tables and querying them with standard SQL, eliminating the need for external vector database infrastructure.

How do I store and query high-dimensional embeddings in SQLite?▼

You store and query high-dimensional embeddings in SQLite by creating a virtual table with a specified dimensional embedding column, then performing nearest neighbor searches using distance functions provided by the sqlite-vec extension.

Does sqlite-vec support large-scale embedding datasets?▼

sqlite-vec supports large-scale embedding datasets by providing partition-based sharding for virtual tables, allowing efficient vector similarity searches and metadata filtering across extensive collections.

What are the limitations of using SQLite for KNN queries?▼

Using SQLite for KNN queries requires loading the sqlite-vec extension into the database connection, and performance depends on native virtual table structures and distance functions rather than external vector database optimizations.

Related Skills

Explore More Software Engineering Skills→
javierhbrjavierhbr

sqlite-vec

Perform vector similarity search and KNN queries within SQLite databases.

Community
Intermediate
devMoezdevMoez
1

chroma

Stores and retrieves document embeddings with similarity search and metadata filtering.

Community
Basic
vikrant-projectvikrant-project
2

faiss

Perform k-NN search and clustering on dense vectors with GPU acceleration.

Community
Advanced
patrykkopycinskipatrykkopycinski
6

vector-search-setup

Configure Elasticsearch indices with dense_vector mappings for kNN and hybrid retrieval.

Community
Intermediate