memory

Maintain MEMORY.md indexes and structured notes for AI agent context.

1|Updated Feb 9, 2026

Search Tags:#context #learning #memory #organization #notes #knowledge-management #memory-index

Ready to run

npx skills add https://github.com/peteknowsai/swain-agents --skill memory-peteknowsai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/peteknowsai/swain-agents/tree/main/sprite/skills/memory
Command: npx skills add https://github.com/peteknowsai/swain-agents --skill memory-peteknowsai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Memory skill provides a structured approach to knowledge management for AI agents, guiding what to remember, when to write, and where to store it so context can be loaded at session start and learning can be organized over time.

Core Features & Use Cases

Memory index management: Maintain a MEMORY.md index that tracks confirmed facts, yearnings, and daily notes.
Lifecycle discipline: Treat yearnings as questions that become memory entries and consolidate them via the dream cycle.
Context loading at session start: Load relevant memory files to calibrate responses and decisions.

Quick Start

Initialize MEMORY.md and populate the memory index with key topics at the start of each session.

Frequently Asked Questions about memory

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

FAQPage Schema

How do I maintain context and memory for AI agents across different sessions?▼

To maintain context for AI agents across sessions, use a structured MEMORY.md index that tracks confirmed facts, yearnings, and notes, allowing relevant memory files to be loaded at session start for rapid recall.

What is the best way to organize knowledge management notes for AI learning?▼

Organizing knowledge management notes for AI learning requires a structured notes directory paired with a frontmatter-driven MEMORY.md index, ensuring consistent memory tracking and efficient context retention over time.

How do I initialize a memory index when starting a new AI session?▼

To initialize a memory index at session start, create a MEMORY.md file and populate it with key topics using frontmatter-based YAML schemas to track confirmed facts, yearnings, and daily notes.

How does the dream cycle consolidation process work for memory indexes?▼

The dream cycle consolidation process treats yearnings as questions that eventually become confirmed memory entries, ensuring organized learning and consistent memory management across ongoing knowledge growth.

Do I need any specific dependencies or frameworks to use memory indexing?▼

No specific dependencies or frameworks are required for memory indexing; the system relies on a frontmatter-driven MEMORY.md index, YAML schemas, and a structured notes directory to function independently.

When should I not use a structured memory index for AI context retention?▼

A structured memory index may be unnecessary for single-session tasks or ephemeral interactions where long-term context retention, ongoing knowledge growth, and cross-session learning are not required.