memory-tiering

Organize multi-tier memory into HOT, WARM, and COLD storage tiers.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill reduces context overload by reorganizing memory into clear HOT, WARM, and COLD tiers so active work stays focused and long-term information remains easy to recover.

Core Features & Use Cases

  • Tiered Memory Management: Keeps immediate session details in HOT, stable preferences and system facts in WARM, and historical summaries in COLD.
  • Compaction Support: Helps after compaction by moving dead context out of the active window while preserving important information.
  • Use Case: When a conversation grows too large, use this Skill to prune completed tasks, retain durable user preferences, and archive decisions into concise memory records.

Quick Start

Use the memory tiering skill to reorganize the current memory layers and remove dead context after compaction.

Frequently Asked Questions about memory-tiering

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

FAQPage Schema
What is memory tiering and how does it manage session context?

Memory tiering organizes multi-tier memory into HOT, WARM, and COLD storage to manage session context. It redistributes live context, stable facts, and historical summaries so active work stays focused and long-term information remains easy to recover.

How do I prune context and preserve important information after compaction?

To prune context after compaction, use memory tiering to move dead context out of the active window. It retains durable user preferences in WARM storage and archives decisions into concise COLD memory records while verifying critical information is preserved.

When should I reorganize memory into HOT, WARM, and COLD tiers?

You should reorganize memory into HOT, WARM, and COLD tiers when a conversation grows too large and causes context overload. This process prunes completed tasks from the active window while keeping immediate session details readily accessible.

Can I archive historical summaries and retain user preferences across ongoing conversations?

Yes, memory tiering archives historical summaries into COLD storage and retains durable user preferences in WARM storage across ongoing conversations. It reads memory tier files and daily logs to redistribute information while verifying critical facts are preserved.

Does context pruning require reading memory tier files and daily logs?

Yes, context pruning requires reading memory tier files and daily logs to accurately redistribute live context, stable facts, and historical summaries. This ensures critical information is preserved during session compaction and archival summarization.

What is the best way to keep context lean without losing stable facts?

The best way to keep context lean is applying a memory tiering strategy that separates immediate session details from stable facts and historical summaries. This prevents context overload while ensuring long-term information remains easy to recover.