030102 lancedb index

Skill natuleadan/skills/skills/030102-lancedb-index

Multi-domain agent skills collection for AI coding agents (Claude, Cursor, Copilot, OpenCode, and more). Covers programming, biology, cooking, and future domains. Installable via npx skills add.

Install

npx -y skills add natuleadan/skills --skill 030102-lancedb-index

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

LanceDB vector index types (IVF, HNSW, PQ, RQ), quantization techniques, reindexing strategies, and performance tuning.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.4 KB, as published. Nobody here has run it

Vector Indexing & Optimization

Overview

Index types, quantization, and reindexing strategies for LanceDB vector search performance.

Quick Reference

Index Types

Index	Best For	Quantization
IVF_PQ	General purpose	Product Quantization
IVF_HNSW_SQ	High recall	Scalar Quantization
IVF_HNSW_PQ	Large datasets	Product Quantization
IVF_FLAT	Small datasets, 100% recall	None
IVF_RQ	Binary vectors	RaBitQ (1 bit/dim)

Quantization

Type	Bits/Dim	Compression	Recall Impact
None	32	1x	None
SQ	8	4x	Minimal
PQ	4-8	4-8x	Moderate
RQ (RaBitQ)	1	32x	Low-moderate

Reindexing

table.optimize()  # Compaction + cleanup + index update

References

Vector Index Types — IVF, HNSW, PQ, RQ index selection and tuning
Indexing Concepts — Internal mechanics, disk-based indexing
Quantization — Compression techniques, accuracy tradeoffs
Reindexing — Incremental updates, compaction, optimization