Memory Bank MCP

NPM Version Semgrep CE scan License: MIT

An MCP server that gives AI assistants persistent memory across sessions. It stores project context, decisions, and progress in structured markdown files — locally or on a remote server via SSH.

Related repos:
HTTP + Postgres + Redis variant → diaz3618/memory-bank-mcp-http — Docker deployment with HTTP transport
VS Code Extension → diaz3618/Memory-Bank-VSCode-Ext — sidebar UI and GitHub Copilot integration

Quick Start

# Run directly (no install needed)
npx @diazstg/memory-bank-mcp

# Or install globally
npm install -g @diazstg/memory-bank-mcp

Via Smithery (Claude Desktop)

npx -y @smithery/cli install @diazstg/memory-bank-mcp --client claude

Configuration

Add to your editor's MCP config (.vscode/mcp.json, Cursor, Claude Desktop, etc.):

{
  "servers": {
    "memory-bank-mcp": {
      "command": "npx",
      "args": ["-y",
          "@diazstg/memory-bank-mcp",
          "--username",
          "your-username"
      ],
      "type": "stdio"
    }
  }
}

Tip: Including --username is highly recommended for proper progress tracking.

Common Options

npx @diazstg/memory-bank-mcp --username "github-user"   # Username for progress tracking (recommended)
npx @diazstg/memory-bank-mcp --mode code                # Set operational mode
npx @diazstg/memory-bank-mcp --path /my/project         # Custom project path
npx @diazstg/memory-bank-mcp --folder my-memory         # Custom folder name (default: memory-bank)
npx @diazstg/memory-bank-mcp --help                     # All options

Remote Server (SSH)

Store your Memory Bank on a remote server:

npx @diazstg/memory-bank-mcp --remote \
  --remote-user username \
  --remote-host example.com \
  --remote-path /home/username/memory-bank \
  --ssh-key ~/.ssh/id_ed25519

See Remote Server Guide.

How It Works

Memory Bank stores project context as markdown files in a memory-bank/ directory:

File	Purpose
`product-context.md`	Project overview, goals, tech stack
`active-context.md`	Current state, ongoing tasks, next steps
`progress.md`	Chronological record of updates
`decision-log.md`	Decisions with context and rationale
`system-patterns.md`	Architecture and code patterns

The AI assistant reads these files at the start of each session and updates them as work progresses, maintaining continuity across conversations.

MCP Tools

Tool	Description
`initialize_memory_bank`	Create a new Memory Bank
`get_memory_bank_status`	Check current status
`read_memory_bank_file`	Read a specific file
`write_memory_bank_file`	Write/update a file
`track_progress`	Add a progress entry
`log_decision`	Record a decision
`update_active_context`	Update current context
`switch_mode`	Change operational mode
`graph_upsert_entity`	Create or update a knowledge graph entity
`graph_add_observation`	Add an observation to an entity
`graph_link_entities`	Create a relation between entities
`graph_search`	Search entities by name or type
`graph_open_nodes`	Get full details of specific entities
`graph_compact`	Compact the event log

Modes

Mode	Focus
`code`	Implementation and development
`architect`	System design and planning
`ask`	Q&A and information retrieval
`debug`	Troubleshooting and diagnostics
`test`	Testing and quality assurance

Modes can be set via CLI (--mode code), tool call (switch_mode), or .mcprules-[mode] files. See Usage Modes.

As a Library

import { MemoryBankServer } from "@diazstg/memory-bank-mcp";

const server = new MemoryBankServer();
server.run().catch(console.error);

Documentation

Topic	Link
Getting Started	npx usage, build with Bun, custom folder
Guides	Remote server, usage modes, status system, debug MCP
Integrations	VS Code/Copilot, Claude Code, Cursor, Cline, Roo Code, generic MCP
Reference	MCP protocol, rules format, file naming
Development	Architecture, testing, logging

Alternative: HTTP + PostgreSQL + Redis

The feature/http-postgres-redis-supabase branch provides a cloud-native variant that replaces stdio/local-filesystem with HTTP Streamable MCP transport, PostgreSQL (via Supabase) for storage, and Redis for caching. It is deployed exclusively via Docker and is not published to npm. See the branch README for setup instructions.