← Back to Skills Marketplace
Memory Tiering
by
SarielWang93
· GitHub ↗
· v1.0.0
16085
Downloads
9
Stars
148
Active Installs
1
Versions
Install in OpenClaw
/install memory-tiering
Description
Automated multi-tiered memory management (HOT, WARM, COLD). Use this skill to organize, prune, and archive context during memory operations or compactions.
Usage Guidance
Install this if you want automated memory cleanup. Before relying on it, consider asking the agent to preview or back up memory changes before pruning, and avoid keeping raw secrets in memory files.
Capability Analysis
Type: OpenClaw Skill
Name: memory-tiering
Version: 1.0.0
The skill bundle is benign. It defines a memory tiering system for an AI agent, instructing it to read and write to internal memory files (`memory/hot/HOT_MEMORY.md`, `memory/warm/WARM_MEMORY.md`, `MEMORY.md`, `memory/YYYY-MM-DD.md`). The instructions in `SKILL.md` are clear, directly related to the stated purpose, and do not contain any prompt injection attempts, malicious execution patterns, data exfiltration, or persistence mechanisms. Notably, an instruction regarding 'credentials in HOT' advises the agent to point to root files rather than storing raw secrets, which is a security-positive directive.
Capability Assessment
Purpose & Capability
The stated purpose is automated HOT/WARM/COLD memory tiering, and the artifact instructions directly match that purpose by reading memory tiers, redistributing context, summarizing, pruning, and archiving.
Instruction Scope
The scope is disclosed in SKILL.md, including all three memory tiers and recent daily logs, but it also triggers automatically after /compact and can remove granular COLD-memory details.
Install Mechanism
The bundle contains only SKILL.md; no executable scripts, install hooks, package dependencies, binaries, or hidden setup behavior were found.
Credentials
Reading memory files and daily logs is proportionate for memory tiering. The credential instruction is security-positive because it advises pointing to root files instead of storing raw secrets.
Persistence & Privilege
The skill intentionally changes durable memory by moving, summarizing, and pruning content. That is expected for the purpose, but it may affect what future sessions remember.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install memory-tiering - After installation, invoke the skill by name or use
/memory-tiering - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
memory-tiering 1.0.0
- Introduces dynamic, three-tiered memory management: HOT, WARM, and COLD.
- Automates context organization, pruning, and archival using a structured workflow.
- Supports both manual and automatic triggers (e.g., after compaction).
- Enhances retrieval efficiency and context relevance by enforcing tier redistribution and summarization steps.
Metadata
Frequently Asked Questions
What is Memory Tiering?
Automated multi-tiered memory management (HOT, WARM, COLD). Use this skill to organize, prune, and archive context during memory operations or compactions. It is an AI Agent Skill for Claude Code / OpenClaw, with 16085 downloads so far.
How do I install Memory Tiering?
Run "/install memory-tiering" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Memory Tiering free?
Yes, Memory Tiering is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Memory Tiering support?
Memory Tiering is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Memory Tiering?
It is built and maintained by SarielWang93 (@sarielwang93); the current version is v1.0.0.
More Skills