memory-manager

Organize local agent memory into structured stores with snapshots and recovery checkpoints.

Updated Feb 16, 2026
One-click install
npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill memory-manager-wewetv1987-cell
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-manager
Source: https://github.com/wewetv1987-cell/openclaw-backup/tree/main/skills/memory-manager
Command: npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill memory-manager-wewetv1987-cell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents context loss by organizing agent memory into a durable structure instead of leaving everything in flat Markdown files. It helps detect when memory is getting too full, preserve important context before compression, and keep long-running work understandable over time.

Core Features & Use Cases

  • Compression Detection: Checks memory usage and flags warning or critical thresholds before context is lost.
  • Snapshots and Recovery: Saves recent episodic, semantic, and procedural context into backup snapshots for safe recovery.
  • Memory Organization: Migrates flat notes into episodic, semantic, and procedural folders for better retrieval.
  • Search and Categorization: Searches memories by type and manually routes content into the right memory store.
  • Operational Monitoring: Tracks counts, size, growth, and health so agents can manage memory proactively.
  • Use Case: An agent that has accumulated days of notes can run this Skill to snapshot important context, reorganize scattered files, and quickly find what happened, what it knows, and how to repeat a workflow.

Quick Start

Ask the assistant to initialize the memory structure, check compression risk, and organize your existing notes into episodic, semantic, and procedural memories.

Frequently Asked Questions about memory-manager

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

FAQPage Schema
How do I prevent context loss in long-running AI agent workflows?

To prevent context loss in long-running AI agent workflows, organize local agent memory into structured stores with recovery checkpoints. This preserves important context before compression and keeps historical data searchable.

How do I organize flat Markdown files into structured agent memory?

Organize flat Markdown files into structured agent memory by migrating scattered notes into episodic, semantic, and procedural folders. This separation enables type-aware search and better retrieval of past context.

When should I use memory compression detection for AI agents?

Use memory compression detection for AI agents when accumulated notes risk exceeding context limits. It tracks memory usage, size, and growth, flagging warning or critical thresholds before context is lost.

Can I backup and recover episodic, semantic, and procedural memory snapshots?

Yes, you can backup and recover episodic, semantic, and procedural memory snapshots. The system saves recent context into backup snapshots, ensuring safe recovery of specific memory types.

Does agent memory organization require shell-based file operations?

Yes, agent memory organization requires shell-based file operations to handle state tracking and memory migration. It safely manages episodic, semantic, procedural, and legacy memory content across Markdown files.

What is the best way to search agent memory history by type?

The best way to search agent memory history by type is using type-aware search across categorized memory stores. This routes content into the right memory folder and quickly retrieves what happened, what is known, and workflow steps.