memory

Manage 5-tier memory and taste data across episodic, semantic, procedural, error-solution, and graph memories.

6|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/waitdeadai/minmaxing --skill memory-waitdeadai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/waitdeadai/minmaxing/tree/main/.claude/skills/memory
Command: npx skills add https://github.com/waitdeadai/minmaxing --skill memory-waitdeadai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This memory skill coordinates a comprehensive 5-tier memory and taste management system, helping keep decisions, memories, and insights aligned across episodic, semantic, procedural, error-solution, and graph memory.

Core Features & Use Cases

  • Centralized 5-tier memory model to log, review, and retrieve memories and taste verdicts.
  • Supports episodic logging (.taste/sessions/*.jsonl), semantic knowledge (taste.md), procedural patterns, error-solution records, and narrative graphs for decision justification.
  • Use Case: You want to log a new episodic memory and immediately review the latest taste decisions to guide next actions.

Quick Start

Log a new episodic memory entry using the provided script to see memory counts and taste insights.

Frequently Asked Questions about memory

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

FAQPage Schema
How do I log and retrieve episodic memory for AI decision tracking?

You log episodic memory using provided scripts that write entries to JSONL session files. This episodic memory tracking records specific events and taste verdicts, allowing you to immediately review recent decisions to guide subsequent actions.

What is a 5-tier memory system and when do I need it?

A 5-tier memory system organizes data across episodic, semantic, procedural, error-solution, and graph memory categories. You need this multi-tier memory model when coordinating the logging, reviewing, and retrieving of diverse decisions and insights from a centralized structure.

How do I review taste decisions and semantic knowledge in a memory log?

You review taste decisions by running scripted commands that list memory stats and extract taste insights. Semantic knowledge is stored in markdown files, allowing you to retrieve logged taste verdicts and semantic memory patterns to inform next actions.

Can I store error-solution records and procedural patterns in a centralized memory model?

Yes, the centralized memory model supports storing error-solution records and procedural patterns. You can log error-solution memory to retain troubleshooting outcomes and record procedural memory to capture recurring patterns for future retrieval and decision justification.

Does this memory system support narrative graphs for decision justification?

Yes, the memory system supports narrative graphs as one of its 5 tiers. Graph memory captures narrative relationships to provide decision justification, allowing you to retrieve contextual connections and review why specific memory-logged choices were made.