memory-admin

Diagnose and manage SurrealDB-backed AI memory with bulk archive and promotion operations.

1|Updated Feb 23, 2026
One-click install
npx skills add https://github.com/baladithyab/engram --skill memory-admin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-admin
Source: https://github.com/baladithyab/engram/tree/main/skills/memory-admin
Command: npx skills add https://github.com/baladithyab/engram --skill memory-admin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the need for proactive management and optimization of an AI's persistent memory, ensuring data integrity, efficient storage, and optimal performance.

Core Features & Use Cases

  • Memory Diagnostics: Provides tools to check the health and status of the memory system, including connection status, memory counts by scope and type, and age of memories.
  • Bulk Operations: Enables archiving stale memories, promoting valuable session memories to project scope, and permanently purging forgotten data.
  • Deployment Mode Management: Allows switching between different memory storage and access modes (embedded, memory, local, remote).
  • Troubleshooting: Offers guidance on common issues like connection errors, empty recall results, slow queries, and record not found errors.

Quick Start

Use the memory-admin skill to check the current memory status.

Frequently Asked Questions about memory-admin

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

FAQPage Schema
How do I check the health and status of my AI memory system?

You can troubleshoot common AI memory issues like connection errors, empty recall results, slow queries, and record not found errors using dedicated diagnostic tools. These tools check connection status and memory counts to identify the root cause.

How do I archive stale memories and promote valuable session data?

Use bulk operations to archive stale memories and promote valuable session memories to project scope. This manages the memory lifecycle by permanently purging forgotten data and optimizing storage based on importance and access counts.

Can I switch between different memory storage and access modes?

Yes, you can switch between different deployment modes for memory storage and access, including embedded, memory, local, and remote. This allows you to optimize your AI's persistent memory performance based on your current infrastructure needs.

What is the best way to maintain a knowledge graph backed by SurrealDB?

Maintain a SurrealDB-backed knowledge graph by using diagnostic tools to monitor memory scopes and bulk operations to archive stale data. Effective administration requires understanding memory importance, access counts, and lifecycle management.

Why does my AI memory return empty recall results or connection errors?

Empty recall results or connection errors in your AI memory often stem from deployment mode misconfigurations or stale data. Use troubleshooting guidance to check connection status, verify memory counts by scope, and resolve record not found errors.