advanced-memory-architect

Optimize Qdrant memory systems for OpenClaw agents with hybrid search, MMR, and temporal decay.

Updated Apr 5, 2026
One-click install
npx skills add https://github.com/Shadojin94/zetamind-app --skill advanced-memory-architect
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: advanced-memory-architect
Source: https://github.com/Shadojin94/zetamind-app/tree/main/skills_openclaw/advanced-memory-architect
Command: npx skills add https://github.com/Shadojin94/zetamind-app --skill advanced-memory-architect

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Optimize Qdrant-based memory systems for OpenClaw agents to improve recall, reduce latency, and balance memory diversity.

Core Features & Use Cases

  • Hybrid search tuning: adjust vectorWeight and textWeight for domain-specific retrieval balance.
  • MMR diversity: enforce non-redundant results while preserving relevance.
  • Temporal decay: configure memory half-lives to favor recent information and aging content.
  • Production readiness: guidelines for profiling, testing, and monitoring memory performance in live deployments.

Quick Start

Configure your memory for openclaw to enable hybrid search, enable MMR, and set temporal decay, then validate results on a representative dataset.

Frequently Asked Questions about advanced-memory-architect

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

FAQPage Schema
How do I tune Qdrant memory for fast and diverse retrieval?

Tune Qdrant memory by configuring hybrid search weights, enabling MMR diversity, and setting temporal decay schedules. This optimizes retrieval pipelines to improve recall, reduce latency, and balance memory diversity for production-grade agent architectures.

What is temporal decay in vector search memory systems?

Temporal decay in vector search memory systems configures memory half-lives to favor recent information and age older content. This mechanism ensures agents retrieve up-to-date context, improving response relevance in adaptive retrieval pipelines.

How do I balance vector and text weights for hybrid search?

Balance vector and text weights for hybrid search by adjusting vectorWeight and textWeight parameters. This tunes domain-specific retrieval balance, allowing you to prioritize semantic similarity or keyword matching based on your specific use case.

Does MMR diversity reduce redundant results in Qdrant retrieval?

MMR diversity reduces redundant results in Qdrant retrieval by enforcing non-redundant results while preserving relevance. This ensures diverse memory recall, preventing duplicate context from overwhelming the agent's working memory.

Can I use Qdrant memory optimization for production deployments?

Qdrant memory optimization supports production deployments by providing guidelines for profiling, testing, and monitoring memory performance. It includes configurable parameters for hybrid weights, MMR settings, decay schedules, and candidate expansion.

Why does my Qdrant memory retrieval return redundant results?

Qdrant memory retrieval returns redundant results when MMR diversity is not configured. Enable MMR settings to enforce non-redundant results while preserving relevance, and adjust candidate expansion parameters to improve overall retrieval quality.