mmw
Enables Codeium (Windsurf) to connect to MMW for persistent memory, semantic search, and memory management capabilities across sessions.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mmwRemember that I prefer concise code examples and TypeScript."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MMW (Managed Memory Workspace) โ MCP Integration & Client Kit
Official Model Context Protocol (MCP) connection kit and client configuration guides for MMW (Managed Memory Workspace).
MMW provides ultra-fast (<2ms) persistent memory, tenant isolation, and cryptographic provenance for autonomous AI agents across sessions and tool executions.
โก Quickstart
MMW supports two connection modes:
๐ก Local-First Resilient Spool (Recommended): Zero data loss. Writes are persisted instantly (<1ms) to local SQLite (
~/.config/mmw/spool.db) and synchronized in the background to MMW Cloud with automatic retries if offline or network fails.๐ Direct Cloud Remote: Connects directly via Streamable HTTP (
https://mcp.mmwhub.tech/mcp).
Related MCP server: Kirok
๐ป Client Configuration
๐ก Mode A: Local-First Resilient Spool (No data loss)
Run via standard Python MCP proxy:
{
"mcpServers": {
"mmw": {
"command": "python3",
"args": ["-m", "mmw_mcp.spool_proxy"],
"env": {
"MMW_TOKEN": "<YOUR_MMW_API_KEY>"
}
}
}
}๐ Mode B: Direct Remote MCP
1. Claude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"mmw": {
"url": "https://mcp.mmwhub.tech/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MMW_API_KEY>"
}
}
}
}2. Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"mmw": {
"url": "https://mcp.mmwhub.tech/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MMW_API_KEY>"
}
}
}
}3. Zed Editor (settings.json)
{
"context_servers": {
"mmw": {
"endpoint": "https://mcp.mmwhub.tech/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MMW_API_KEY>"
}
}
}
}4. Windsurf (~/.codeium/windsurf/mcp_config.json)
{
"mcpServers": {
"mmw": {
"serverUrl": "https://mcp.mmwhub.tech/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MMW_API_KEY>"
}
}
}
}๐ Available MCP Tools
Tool | Description |
| Store long-term knowledge with confidence score, source ID, hash, and provenance tags. |
| Instant (<2ms) semantic retrieval over past session decisions, facts, and constraints. |
| Validate memory consistency, verify facts, and resolve contradictory statements. |
| Prune or mark obsolete memories as superseded. |
| Execute tenant-isolated external integrations and connected project tools. |
๐ One-Click Setup Script
We provide automated setup scripts in this repository for macOS, Linux, and Windows:
# macOS / Linux
chmod +x connect.sh
MMW_CREDENTIAL='<YOUR_MMW_API_KEY>' ./connect.sh --client cursor --verify
# Windows (PowerShell)
$env:MMW_CREDENTIAL='<YOUR_MMW_API_KEY>'
.\connect.ps1 -Client cursor -Verify๐ Links & Resources
Website: https://mmwhub.tech
Web App & API Keys: https://app.mmwhub.tech
Smithery Registry: https://smithery.ai/server/mmsingatullin/mmw
Documentation: https://mmwhub.tech/docs
This server cannot be deployed
Maintenance
Related MCP Connectors
Persistent memory for AI agents. EU-hosted, privacy-first, hybrid recall, contradiction detection.
Hosted persistent memory with semantic search, importance and TTL for AI agents.
Document hosting and encrypted agent memory with multi-tenant persistence.
Universal persistent memory and knowledge retrieval layer for AI agents and LLMs.
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables AI agents with persistent semantic memory, including semantic recall, knowledge graphs, and instant domain expertise via pre-built Intelligence Packs.1027 npmMIT
- AlicenseAqualityAmaintenanceProvides persistent, searchable memory for AI agents, enabling them to retain, recall, and reflect on information across conversations.1919 PyPI1MIT
- AlicenseNot gradedqualityCmaintenanceGives AI agents persistent memory, handoffs, and shared context across sessions, enabling seamless continuity and multi-agent collaboration.65 npm68-
- AlicenseCqualityAmaintenanceEnables AI agents to maintain persistent, searchable two-layer memory with 37 tools, hybrid search, knowledge graphs, and enterprise features like authentication and backups.641 PyPIMIT