cog-housekeeping

Automate weekly memory pruning, archival, and glacier index regeneration.

Updated Apr 24, 2026
One-click install
npx skills add https://github.com/bketelsen/max --skill cog-housekeeping
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cog-housekeeping
Source: https://github.com/bketelsen/max/tree/main/skills/cog-housekeeping
Command: npx skills add https://github.com/bketelsen/max --skill cog-housekeeping

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates long-running memory maintenance tasks to keep Max's cognitive store organized, efficient, and within performance limits.

Core Features & Use Cases

  • Weekly housekeeping: prune hot-memory.md, archive stale observations, and rebuild glacier/index.md.
  • Glacier & index maintenance: move old data into memory/glacier and regenerate the glacier index for fast recovery.
  • Link audit & health: audit inter-domain references and ensure links remain consistent.

Quick Start

Tell Max to run cog-housekeeping to prune memory, archive data, and rebuild the glacier index.

Frequently Asked Questions about cog-housekeeping

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

FAQPage Schema
How do I automate memory housekeeping and archive stale data?

Glacier indexing moves old data into a memory/glacier archive and regenerates the glacier index for fast recovery. This process keeps hot memory lean and ensures consistent discovery and retrieval of archived memory-mapped domains.

What is the best way to audit inter-domain memory references?

Audit inter-domain memory references by running an automated link audit to ensure links remain consistent across memory-mapped domains. This health check verifies that inter-domain references stay valid after pruning and archiving operations.

Why does hot memory become slow and how do I prune it?

Hot memory becomes slow when stale observations accumulate and exceed performance limits. Prune hot memory by automating memory housekeeping tasks that archive old data to glacier storage and regenerate the YAML frontmatter metadata index.

Does memory pruning require YAML frontmatter metadata?

Yes, memory pruning requires YAML frontmatter metadata to fulfill domain-specific glacier rules and index regeneration workflows. This metadata ensures consistent discovery and retrieval after moving old data into the glacier archive.

When should I rebuild the glacier index for memory-mapped domains?

Rebuild the glacier index during weekly housekeeping or after archiving stale observations to ensure consistent discovery and retrieval. Regenerating the index keeps memory-mapped domains healthy and supports fast recovery of archived data.