memory

Maintain long-term facts and conversation history across sessions using MEMORY.md and HISTORY.md.

3|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/matparang/AutoJaga --skill memory-matparang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/matparang/AutoJaga/tree/main/legacy/jagabot/skills/memory
Command: npx skills add https://github.com/matparang/AutoJaga --skill memory-matparang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Maintain long-term facts and conversation context across sessions to preserve continuity and reduce repetition.

Core Features & Use Cases

  • Two-layer memory: MEMORY.md stores long-term facts and project context, loaded into your active context.
  • History logging: HISTORY.md acts as an append-only log for searchable events via grep.
  • Auto-consolidation: Conversations are summarized and merged into MEMORY.md to keep knowledge up to date.

Quick Start

Store a fact or memory item and retrieve it later.

Frequently Asked Questions about memory

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

FAQPage Schema
How do I maintain long-term memory and context across multiple sessions?

Long-term memory is maintained using a two-layer design: MEMORY.md stores persistent facts loaded into active context, while HISTORY.md logs append-only conversation events for searchable recall.

What's the best way to recall past user preferences and project history?

Recall past preferences and project history by using grep-based search on an append-only HISTORY.md log file, which records events and context for ongoing work.

How does automatic consolidation work for conversation history?

Automatic consolidation summarizes ongoing conversations and merges the key facts into MEMORY.md, ensuring long-term knowledge remains updated without manual editing.

Can I use persistent memory for ongoing collaborative projects?

Persistent memory is ideal for ongoing projects and collaborative work where recalling past events and preserving context continuity across sessions is required.

Do I need any specific dependencies to set up long-term memory?

No specific dependencies are required to set up long-term memory; the system operates using standard markdown files loaded directly into your active context.

Why does my context reset between sessions instead of keeping history?

Context resets because session data is volatile; storing facts in MEMORY.md and logging events to HISTORY.md solves this by providing persistent, grep-searchable long-term memory.