Memory Lancedb Pro

This is a serious upgrade over basic vector search for Claude's memory. It combines vector similarity with BM25 keyword search, fuses results with reciprocal rank fusion, and optionally reranks with a cross-encoder for better relevance. The multi-scope isolation is smart if you're running multiple agents or projects and don't want memory bleed. It auto-captures facts and preferences during conversation, filters out noise like greetings and emoji spam, and ships with a CLI for bulk operations and migration from simpler backends. The architecture is clean, embedding providers are swappable, and the scoring pipeline is well-documented if you need to tweak retrieval behavior. Worth it if memory quality matters to your use case.

skills.sh installs: 645
GitHub stars: 47
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add win4r/memory-lancedb-pro-skill --skill memory-lancedb-pro --agent claude-code

Installs into .claude/skills of the current project.

Files

memory-lancedb-pro Plugin Maintenance Guide

Overview

memory-lancedb-pro is an enhanced long-term memory plugin for OpenClaw. It replaces the built-in memory-lancedb plugin with advanced retrieval capabilities, multi-scope memory isolation, and a management CLI.

Repository: https://github.com/win4r/memory-lancedb-pro License: MIT | Language: TypeScript (ESM) | Runtime: Node.js via OpenClaw Gateway