skills.rest
Nick-Tsai
Nick-TsaiCOMMUNITY
Explore all skills by this author →
GitHub

memory-system

Manage persistent AI agent memory with Markdown files and Git synchronization.

1|Updated Mar 16, 2026
Search Tags:#structured data#ai memory#git sync#document-driven#first-person perspective#twin mode
One-click install
npx skills add https://github.com/Nick-Tsai/twin-memory --skill memory-system-nick-tsai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-system
Source: https://github.com/Nick-Tsai/twin-memory/tree/main
Command: npx skills add https://github.com/Nick-Tsai/twin-memory --skill memory-system-nick-tsai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a persistent, structured, and human-readable memory system for AI agents, enabling them to recall past interactions, learn from experiences, and maintain a consistent identity across different instances (cloud and physical devices).

Core Features & Use Cases

  • Structured Memory: Records interactions using Context, Facts, and Feelings dimensions.
  • Durable Storage: Uses Markdown files with YAML frontmatter, making memories human-readable and version-control friendly.
  • Twin Mode Synchronization: Enables cloud and physical AI instances to share and synchronize memories via Git.
  • Layered Context Loading: Optimizes AI context by prioritizing essential metadata and recent memories.
  • Use Case: An AI assistant can use this system to remember user preferences, past conversations, and its own evolving understanding of the world, ensuring continuity and personalization across sessions and devices.

Quick Start

Use the memory-system skill to add a new memory about a recent conversation.

Frequently Asked Questions about memory-system

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

FAQPage Schema
How do I create a persistent memory system for an AI assistant across different sessions?▼

Twin mode synchronization enables cloud and physical AI instances to share memories via Git synchronization. By pushing and pulling Markdown memory files through a Git repository, different devices can maintain a unified context and consistent identity.

Does Git synchronization work for sharing AI memory between cloud and physical instances?▼

Git synchronization enables cloud and physical AI instances to share a persistent memory system via twin mode. By pushing and pulling Markdown files through a repository, different devices maintain a unified context and consistent identity across sessions.

What is the best way to structure AI memory data for efficient context loading?▼

Structured AI memory uses multi-dimensional recording across Context, Facts, and Feelings dimensions, paired with layered context loading. This mechanism prioritizes essential metadata and recent memories to optimize recall without exceeding token limits.

Can I store AI agent memories in Markdown files instead of a database?▼

You can store AI agent memories in Markdown files with YAML frontmatter instead of a database. This document-driven approach keeps memories human-readable, easy to edit manually, and version-control friendly for Git synchronization.

What dimensions should I record when building a structured memory for AI agents?▼

When building structured memory for AI agents, you should record Context, Facts, and Feelings dimensions. This multi-dimensional approach captures the conversation history, objective information, and the AI's evolving understanding of the world.

Do I need Git to use a document-driven memory system for AI?▼

Git is not strictly required to use a document-driven memory system locally, but it is necessary for twin mode synchronization. Git enables cloud and physical AI instances to share and synchronize Markdown memory files across devices.

Related Skills

Explore More Software Engineering Skills→
Tx1207Tx1207
5

project-creator

Create a new project from a template, initialize uv, and configure a local Git repository.

Community
Advanced
MiniMax-AIMiniMax-AI
2.9k

skill-creator

Generate Skill Units with YAML frontmatter, Markdown body, and bundled resources.

Official
Intermediate
ChandlerHardyChandlerHardy
1

chronicle-remote-summarizer

Export remote Chronicle sessions and summarize them locally via the Gemini API.

Community
Advanced
jonnabiojonnabio
6

chroma

Index and retrieve documents using semantic embeddings and metadata.

Community
Intermediate