mb-init

Initialize Memory Bank directories, docs, and command proxy skills in a repository.

53|11|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/mrvladd-d/memobank --skill mb-init
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mb-init
Source: https://github.com/mrvladd-d/memobank/tree/main/skills/mb-init
Command: npx skills add https://github.com/mrvladd-d/memobank --skill mb-init

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

mb-init creates a ready-to-use Memory Bank skeleton in the current repository, providing the base structure and proxies to start using memobank workflows.

Core Features & Use Cases

  • Initialize essential Memory Bank structure: .memory-bank/, .tasks/, .protocols/, AGENTS.md, CLAUDE.md link/copy to proxies
  • Seed command proxy scaffolding for each runtime: create native skills under .claude/skills and .agents/skills
  • Seed MB router/index docs and command stubs to support immediate operation

Quick Start

Run mb-init to seed the Memory Bank skeleton and native proxy skills into your repository.

Frequently Asked Questions about mb-init

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

FAQPage Schema
How do I initialize a memory bank skeleton in my repository?

You can initialize a memory bank skeleton by running mb-init, which automatically creates the .memory-bank/, .tasks/, and .protocols/ directories while seeding essential docs like AGENTS.md and CLAUDE.md for immediate workflow operation.

What is the purpose of seeding command proxy scaffolding for AI agents?

Seeding command proxy scaffolding creates native skills under .claude/skills and .agents/skills so that different AI runtimes can directly execute memory-bank workflows without manual configuration.

Can I use memory-bank workflows with both Claude and generic agent runtimes?

Yes, memory-bank workflows support multiple runtimes by generating native proxy skills in both .claude/skills and .agents/skills directories, enabling execution across different agent environments.

What's the best way to bootstrap project memory and task tracking directories?

The best way to bootstrap project memory and task tracking directories is using an initialization skill that generates the .memory-bank/ and .tasks/ folders, pre-seeded with router index docs and command stubs.

Do I need to manually configure AGENTS.md and CLAUDE.md to start memory-bank workflows?

No, you do not need to manually configure AGENTS.md and CLAUDE.md; the initialization process automatically seeds these essential docs and links them to the generated command proxies.