skills.rest
papachong
papachongCOMMUNITY
Explore all skills by this author →
GitHub

memory-manager

Initialize three-tier agent memory and detect compression risk with snapshots.

2|Updated Apr 4, 2026
Search Tags:#agents#memory#compression#snapshot#semantic-memory#episodic-memory#procedural-memory
One-click install
npx skills add https://github.com/papachong/RHClaw --skill memory-manager-papachong
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-manager
Source: https://github.com/papachong/RHClaw/tree/main/RHClaw-Desktop/src-tauri/skills/memory-manager
Command: npx skills add https://github.com/papachong/RHClaw --skill memory-manager-papachong

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq.

What problem does it solve?

Agents and local agent systems lose context when memory grows unstructured or approaches storage limits; this Skill organizes agent memory into episodic, semantic, and procedural tiers and detects compression risk to prevent context loss.

Core Features & Use Cases

  • Initializes a three-tier local memory structure (episodic, semantic, procedural) and snapshots for safe backups.
  • Detects compression risk with configurable thresholds, creates recovery snapshots, and provides statistics and growth monitoring.
  • Organizes flat notes into the proper tiers, supports manual categorization, and enables targeted search by memory type for agent context retrieval.
  • Use Case: Run periodic checks on an agent runtime to auto-snapshot critical knowledge and migrate legacy notes into structured memory so agents retain meaningful context during long-running interactions.

Quick Start

Use the memory-manager to initialize your workspace, run a compression check, and create a snapshot if needed.

Frequently Asked Questions about memory-manager

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

FAQPage Schema
How do I organize local agent memory to prevent context loss during long-running interactions?▼

You can structure agent memory into episodic, semantic, and procedural tiers to prevent context loss. This approach categorizes event logs, factual knowledge, and task procedures separately, enabling targeted search and retrieval during long-running interactions.

How does agent memory compression detection work with configurable thresholds?▼

Memory compression detection scans memory files against configurable size thresholds to identify compression risk. When risk is detected, the system triggers automated snapshot creation for backup and provides statistics for ongoing growth monitoring.

What is the best way to migrate flat notes into structured agent memory tiers?▼

Migrate flat notes into structured agent memory by using manual categorization utilities to move unstructured files into episodic, semantic, or procedural tiers. This organizes legacy notes so agents can search memories by type.

Does the memory manager require jq to run in a local Unix workspace?▼

The memory manager operates in a local Unix or OpenClaw workspace using basic JSON state tracking, with jq providing optional support for advanced JSON parsing. You can initialize memory directories and run compression checks without it.

When should I create snapshots for agent memory recovery?▼

Create snapshots for agent memory recovery when compression checks detect risk near storage limits or before migrating flat notes. The system supports automated snapshot creation during periodic checks to ensure critical knowledge is safely backed up.

Related Skills

Explore More Software Engineering Skills→
MakisuoMakisuo
1.6k

clickhousectl-local-dev

Install and configure a local ClickHouse environment with clickhousectl.

Community
Intermediate
melikhanmutlumelikhanmutlu

agent-memory-systems

Structure agent memory with short-term, long-term, and retrieval strategies.

Community
Advanced
Wike-CHIWike-CHI
2

memory-manager

Detect memory compression and create snapshots in local agent workspaces.

Community
Advanced
kunwl123456kunwl123456
26

memory-manager

Initialize, organize, and search local agent memory by type.

Community
Intermediate