memory-manager

Store persistent project memory in .claude/memory with git versioning.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/arroyo2009jose-max/civradar-autoridades --skill memory-manager-arroyo2009jose-max
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-manager
Source: https://github.com/arroyo2009jose-max/civradar-autoridades/tree/main/.claude/skills/memory-manager
Command: npx skills add https://github.com/arroyo2009jose-max/civradar-autoridades --skill memory-manager-arroyo2009jose-max

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Sistema de memoria persistente POR PROYECTO. Guarda conocimiento en .claude/memory/ dentro del repo, versionado con git y compartido con el equipo. Reemplaza la auto-memory de Claude Code que es local y no viaja con el proyecto. Usar PROACTIVAMENTE al inicio de sesion para cargar contexto, cuando el usuario corrige algo, cuando se descubre un patron nuevo, cuando se resuelve un bug, o cuando el usuario dice recordar/recuerda/guarda esto/no olvides. Triggers: recuerda, remember, guarda esto, no olvides, te acuerdas, recuerdas, memoria, en que quedamos, contexto anterior, que sabes de, conversacion pasada, sesion anterior.

Core Features & Use Cases

  • Memoria por proyecto alojada en el repositorio y versionada con git para compartir conocimiento entre colaboradores.
  • Desactiva la memoria automática de Claude Code y la reemplaza con un mecanismo guiado que reside dentro del repositorio.
  • Estructura de carpetas: .claude/memory/{user,feedback,project,reference} y un índice MEMORY.md para navegación y recuperación de información.

Quick Start

Ejecute los pasos de activación inicial para deshabilitar la auto-memoria de Claude Code y crear la estructura .claude/memory, para empezar a guardar el conocimiento del proyecto.

Frequently Asked Questions about memory-manager

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

FAQPage Schema
How do I persist project memory inside a git repository for Claude Code?

To persist project memory inside a git repository, this Skill stores knowledge under the .claude/memory directory structure, versioning it with git so team members can share context across sessions.

How does per-project memory differ from Claude Code's automatic local memory?

Per-project memory resides inside the repository and is versioned with git, allowing team-wide knowledge sharing. It replaces automatic local memory by disabling it and requiring explicit setup via a MEMORY.md index and folder structure.

How do I set up the .claude/memory folder structure for my project?

Setting up the .claude/memory structure requires running initial activation steps that create subdirectories for user, feedback, project, and reference data, alongside a MEMORY.md index for navigation and retrieval.

When should I use repository-based memory management for Claude Code?

Use repository-based memory management during onboarding, bug fixes, or design decisions. It is triggered proactively at session start to load context, when discovering patterns, or when a user corrects information.

Can I share Claude Code context and knowledge across different team members?

Yes, you can share Claude Code context across team members by storing per-project knowledge inside the repository. Because this memory is versioned with git, all collaborators retrieve the same contextual data.

What are the limitations of using explicit project memory instead of auto-memory?

The limitation of explicit project memory is that auto-memory is disabled, requiring manual setup of the .claude/memory structure and a MEMORY.md index. Knowledge is not captured automatically and needs guided triggers to save.