skills.rest
marcoapereirav-arch
marcoapereirav-archCOMMUNITY
Explore all skills by this author →
GitHub

memory-manager

Store persistent project memory in .claude/memory/ with a MEMORY.md index.

Updated Aug 23, 2026
Search Tags:#context#claude#repository#memory#persistence#project#memory-manager
One-click install
npx skills add https://github.com/marcoapereirav-arch/nvision-saas-factory --skill memory-manager-marcoapereirav-arch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-manager
Source: https://github.com/marcoapereirav-arch/nvision-saas-factory/tree/main/saas-factory/.claude/skills/memory-manager
Command: npx skills add https://github.com/marcoapereirav-arch/nvision-saas-factory --skill memory-manager-marcoapereirav-arch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Persistent, per-project memory stored inside the repository, versioned with git, to share knowledge across the team and replace local, non-portable auto-memory.

Core Features & Use Cases

  • Project-scoped memory store in .claude/memory/ with an index MEMORY.md that organizes user, project, feedback, and reference sections.
  • First-run setup automates deactivating Claude Code auto-memory and initializes the memory folders and index.
  • Operational guidance includes clear triggers for reading memory at session start, when topics recur, or before proposing architecture, and how to update MEMORY.md and detail files.
  • Versioned collaboration ensures memory travels with the code and can be reviewed via git history.

Quick Start

Activate memory manager at the start of a session to load MEMORY.md and initialize the .claude/memory structure.

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?▼

You can persist project memory inside a git repository by storing context in a MEMORY.md index and the .claude/memory/ directory, keeping a shared context across sessions versioned with your code.

Can I disable Claude Code auto-memory and replace it with repository-scoped memory?▼

Yes, this approach disables Claude Code auto-memory during a first-run setup and replaces it with a portable, project-scoped memory store initialized in the .claude/memory/ directory.

What is the best way to share development context across a team using git?▼

The best way to share development context is using versioned collaboration, where project memory travels with the code and can be reviewed via git history, ensuring knowledge is shared across the team.

When should I read project memory during an ongoing development session?▼

You should read project memory at session start, when topics recur, or before proposing architecture, using operational guidance triggers to load relevant context from the MEMORY.md index.

Does this project memory approach work for logging decisions and bug fixes?▼

Yes, this project memory approach works for decision logging and bug fixes by organizing context into user, project, feedback, and reference sections within the repository's memory directory.

Related Skills

Explore More Software Engineering Skills→
ramonclaudioramonclaudio
4

add

Clone a GitHub repository into ~/Developer/refs/ and index it with QMD.

Community
Advanced
ten-builderten-builder
106

session-pack

Automate memory capture and handoff planning for collaborative sessions.

Official
Advanced
JKHeadleyJKHeadley
75

instar-identity

Restore persistent agent identity from .instar files after session restarts.

Community
Advanced
Kit4SomeKit4Some
4

session-tracker

Track cross-session context and persist memory across Claude Code conversations.

Community
Advanced