skills.rest
HelloRuru
HelloRuruCOMMUNITY
Explore all skills by this author →
GitHub

memory-engine

Capture decisions, checkpoints, pitfalls, and corrections into structured markdown memory.

132|27|Updated Mar 7, 2026
Search Tags:#markdown#claude code#memory management#project context#session handoff#learning loop#pitfall detection
One-click install
npx skills add https://github.com/HelloRuru/claude-memory-engine --skill memory-engine-helloruru
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-engine
Source: https://github.com/HelloRuru/claude-memory-engine/tree/main/skill
Command: npx skills add https://github.com/HelloRuru/claude-memory-engine --skill memory-engine-helloruru

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It prevents Claude Code from repeating the same mistakes and losing project-specific knowledge between conversations by turning errors and decisions into reusable memory.

Core Features & Use Cases

  • Smart Context: automatically loads the right project's memory based on your working directory, so Claude resumes with the correct context.
  • Student Loop Learning: runs an 8-step learning cycle that checkpoints mid-session, reviews recent outcomes, and refines what to keep or remove.
  • Auto Learn + Correction Cycle: detects pitfalls (retries, errors then fixes, user corrections) and records resolved problems and fixes, while also upgrading rules when the same mistake reappears.

Quick Start

In Claude Code, run the handoff and reflection commands to continue learning across sessions and ensure the latest memory updates are incorporated.

Frequently Asked Questions about memory-engine

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

FAQPage Schema
How do I persist Claude Code project context across multiple sessions?▼

You can persist Claude Code project context across sessions by capturing decisions, checkpoints, and pitfalls into structured markdown memory stored on disk, allowing the AI to resume with the correct background information.

Why does Claude Code keep repeating the same tool-edit mistakes during long-running development?▼

Claude Code repeats tool-edit mistakes when it lacks retained memory of past errors. An auto-correction cycle detects retries and user corrections, recording resolved problems to prevent identical mistakes in future workflows.

How do I hand off a coding session to a new Claude Code window without losing progress?▼

You can hand off a coding session by running a dedicated handoff command that saves current progress and context into markdown memory, ensuring the new window retrieves and continues the exact development state.

Can I manage separate Claude Code memories when switching projects by directory?▼

Yes, smart context loading automatically detects your working directory and loads the corresponding project-specific memory, ensuring Claude Code applies the correct context for each distinct project.

What is a student loop learning cycle for AI memory management?▼

A student loop learning cycle is an 8-step process that checkpoints mid-session, reviews recent outcomes, and refines stored markdown memory rules to continuously improve what the AI retains or discards.

Does this memory management approach require any external dependencies?▼

No, this approach requires zero external dependencies, relying entirely on on-disk markdown persistence and native session lifecycle hooks within the Claude Code environment to function.

Related Skills

Explore More Software Engineering Skills→
cdrgurucdrguru
1

claude-code-init

Initialize Claude Code sessions with PACK context and session logs.

Community
Intermediate
ResgridResgrid
224

session-management

Load and persist .NET project context across AI engineering sessions.

Official
Advanced
kangarookingkangarooking
154

context-management

Manage token budgets and context windows for long-running AI conversations.

Community
Advanced
wilsonfaustinowilsonfaustino
2

wrap-up

Automate end-of-session knowledge capture with Obsidian notes and memory updates.

Community
Intermediate