agent-memory-setup

Configure a 3-tier OpenClaw memory system with HOT/WARM/COLD storage and daily logs.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/revan710517539/SuperTeams --skill agent-memory-setup
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-memory-setup
Source: https://github.com/revan710517539/SuperTeams/tree/main/skills/memory-setup
Command: npx skills add https://github.com/revan710517539/SuperTeams --skill agent-memory-setup

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Create and configure a complete 3-tier OpenClaw memory system for any agent, including HOT/WARM/COLD storage, daily logs, semantic search, and lossless context management, so onboarding and fresh deployments start with a reliable memory foundation.

Core Features & Use Cases

  • 3-tier memory structure with HOT, WARM, and COLD repositories and the corresponding files (HOT_MEMORY.md, WARM_MEMORY.md, MEMORY.md, and daily memory/YYYY-MM-DD.md) to preserve context across sessions.
  • Local semantic search (QMD) over MEMORY.md and memory/*.md to retrieve relevant memories by meaning.
  • Lossless Claw integration to compact long conversations into expandable summaries, reducing amnesia risk.
  • AGENTS.md template provisioning and ready-to-run openclaw.json configuration for memorySearch, compaction, context pruning, and heartbeats.
  • Step-by-step setup script that initializes directories, creates default files, and guides integration of required plugins.

Quick Start

Run the provided setup script in your agent workspace to initialize the memory system.

Frequently Asked Questions about agent-memory-setup

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

FAQPage Schema
How do I set up a 3-tier memory system for a new OpenClaw agent?

To set up a 3-tier memory system, run the provided setup script in your agent workspace to initialize HOT, WARM, and COLD storage directories, generate daily logs, and enable required semantic search plugins.

What is memory tiering and how does it preserve context across sessions?

Memory tiering preserves context by structuring data into HOT, WARM, and COLD repositories with specific markdown files, ensuring important details are retained and searchable while reducing amnesia risk across sessions.

How do I configure semantic search over markdown memory files?

Configure semantic search by enabling the QMD plugin in your openclaw.json file, which allows the agent to retrieve relevant memories by meaning from MEMORY.md and daily log markdown files.

Does this memory setup integrate with lossless context compaction?

Yes, the setup integrates with lossless Claw compaction to compact long conversations into expandable summaries, automatically configuring context pruning and heartbeat settings in openclaw.json.

Can I use this configuration for fresh agent deployments and onboarding?

Yes, this deterministic setup workflow is designed for onboarding and fresh deployments, providing required AGENTS.md templates and default files to establish a reliable memory foundation immediately.

What's the best way to automate OpenClaw memory directory initialization?

Automate memory directory initialization by executing the step-by-step setup script, which creates default files, establishes the memory structure, and guides the integration of required plugins automatically.