skills.rest
Kosmio
KosmioOFFICIAL
Explore all skills by this author →
WebsiteGitHub

memory-management

Implement a three-tier HOT, WARM, COLD memory system for cross-session SEO campaign context.

Updated Apr 10, 2026
Search Tags:#archival#cross-session#memory-management#memory-architecture#seo-memory#hot-cache
One-click install
npx skills add https://github.com/Kosmio/skeleton-test --skill memory-management-kosmio
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-management
Source: https://github.com/Kosmio/skeleton-test/tree/main/.claude/skills/seo/seo-skills/cross-cutting/memory-management
Command: npx skills add https://github.com/Kosmio/skeleton-test --skill memory-management-kosmio

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Maintains cross-session SEO campaign context by organizing data in a three-tier memory model (HOT, WARM, COLD), ensuring critical findings survive across sessions and workflows.

Core Features & Use Cases

  • HOT Cache: memory/hot-cache.md (80 lines max) loaded automatically every session via SessionStart.
  • WARM Storage: dated findings stored under memory/ subdirectories and loaded on demand by other skills.
  • COLD Archive: moves stale data to memory/archive with date prefixes.
  • Cross-Skill Aggregation: aggregates insights across memory/ subdirectories to support multi-skill decisions. Typical use cases include long-running campaigns, post-audit analyses, and cross-skill coordination where memory retention is needed for continuity.

Quick Start

Initialize memory structure for a new SEO project using the provided CLAUDE.md templates and memory folders.

Frequently Asked Questions about memory-management

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

FAQPage Schema
How do I preserve SEO campaign context across multiple sessions?▼

You can preserve cross-session SEO campaign context by implementing a three-tier memory system (HOT, WARM, COLD) that organizes data in a dedicated directory structure, ensuring critical findings and decisions survive across workflows.

What is the best way to manage long-running SEO project memory and avoid losing findings?▼

Managing long-running SEO project memory requires a three-tier architecture: a HOT cache loaded every session, WARM storage for on-demand dated findings, and a COLD archive for stale data with date prefixes.

How does cross-skill memory aggregation work for ongoing SEO analyses?▼

Cross-skill memory aggregation works by collecting insights across memory subdirectories, allowing different SEO skills to access shared findings and support multi-skill decisions during ongoing campaign workflows.

Do I need a specific directory structure to maintain cross-session memory for my campaigns?▼

Yes, maintaining cross-session memory requires a defined directory structure under a memory folder, CLAUDE.md frontmatter with hot-cache configuration, and integration hooks to trigger memory updates after analyses.

When should I move stale SEO findings to a COLD archive?▼

You should move stale SEO findings to a COLD archive when the data is no longer actively needed for current workflows, using date prefixes in the memory archive to maintain organized historical records.

Can I automatically load critical SEO context at the start of a new session?▼

Yes, you can automatically load critical SEO context by maintaining a HOT cache file limited to 80 lines, which is loaded automatically every session via SessionStart to ensure immediate context availability.

Related Skills

Explore More Data & Analytics Skills→
wanghaishengwanghaisheng
10

memory-management

Initialize and maintain a two-layer SEO memory system with CLAUDE.md and memory/ archives.

Community
Advanced
ShiienShiien
7

memory-write

Persist insights and decisions to memory episodes with deduplication and indexing.

Community
Intermediate
zhangzhang-111-izhangzhang-111-i
2

marketing-ops

Route marketing inquiries to specialized skills and orchestrate multi-skill campaigns.

Community
Advanced
shuyu-labsshuyu-labs
288

planning-with-files

Automate complex task planning using persistent Markdown files as working memory.

Official
Basic