Memory

Create categorized long-term memory storage with per-category INDEX.md files.

1|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/zhanbei1/OpenOcta --skill memory-zhanbei1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Memory
Source: https://github.com/zhanbei1/OpenOcta/tree/main/src/skills/memory
Command: npx skills add https://github.com/zhanbei1/OpenOcta --skill memory-zhanbei1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Infinite organizational memory is needed to extend the built-in agent memory with scalable, categorized storage, preventing memory fragmentation and data loss over time.

Core Features & Use Cases

  • Infinite categorized storage under ~/memory/ that runs in parallel with built-in memory.
  • Category-based indexing with per-category INDEX.md and optional syncing from built-in memory.
  • Use cases include long-term project histories, knowledge bases, and decision trails that expand without bound.

Quick Start

Create ~/memory/ with your chosen categories and add your first memory item to verify the system is live.

Frequently Asked Questions about Memory

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

FAQPage Schema
How do I organize long-term AI agent memory to prevent data fragmentation?

Organize long-term AI agent memory by creating a root ~/memory structure with infinite categorized storage. This prevents memory fragmentation and data loss by running parallel to built-in memory.

How do I create an index for categorized AI memory?

Create an index for categorized AI memory by adding per-category INDEX.md files within your ~/memory directory. This category-based indexing structures project histories and knowledge bases effectively.

Can I sync built-in agent memory with a separate long-term storage layer?

Yes, you can optionally sync built-in agent memory with your categorized long-term storage. This synchronization expands decision trails and knowledge bases without bound while maintaining separation.

What is the best way to store expanding project histories for AI agents?

The best way to store expanding project histories is using an infinite categorized memory layer under ~/memory/. This approach scales structured decision trails and histories beyond built-in agent limits.

Do I need a specific framework to set up scalable AI memory?

No specific framework is needed to set up scalable AI memory. You simply create the ~/memory directory with your chosen categories and add your first memory item to verify the system is live.