memory-management

Store and retrieve persistent AI memory using tiered Markdown files.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the challenge of AI context loss by enabling persistent storage and retrieval of crucial information, ensuring continuity and personalized interactions across sessions.

Core Features & Use Cases

  • Tiered Memory Architecture: Utilizes a hot cache (CLAUDE.md) for frequent terms and deep storage (memory/) for comprehensive data.
  • Contextual Decoding: Transforms shorthand into full understanding by looking up terms, people, and projects.
  • Use Case: When a user says "ask todd about the PSR for oracle," the skill decodes "todd" to "Todd Martinez, Finance lead," "PSR" to "Pipeline Status Report," and "oracle" to "Oracle Systems deal," enabling precise action.

Quick Start

Use the memory management skill to add the term 'OKR' meaning 'Objectives & Key Results' to the glossary.

Frequently Asked Questions about memory-management

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

FAQPage Schema
How do I persist context and glossary terms for AI assistant interactions across sessions?

You can persist context by storing glossary terms, people, and project details in structured Markdown files. This enables an AI assistant to recall specific facts and decode shorthand across multiple sessions.

How does tiered memory storage work for AI context management?

Tiered memory uses a hot cache for frequent terms and deep storage directories for comprehensive data. This architecture ensures rapid retrieval of active context while maintaining extensive background knowledge.

How do I decode project codenames and acronyms into full context for an AI agent?

Decoding acronyms involves looking up shorthand against stored glossary files. The system matches codenames to full definitions, enabling precise AI understanding of specialized project vocabulary.

Can I use Markdown files to manage persistent knowledge for my AI assistant?

Yes, structured Markdown files are required to manage persistent knowledge. You must organize context into specific Markdown files for glossary, people, projects, and company context.

What is the best way to maintain personalized AI interactions without losing company context?

Maintaining personalized AI interactions requires storing facts and preferences in a persistent memory system. This saves crucial company context, ensuring continuity and preventing AI context loss.