Personal AI Memory Hub
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., "@Personal AI Memory Hubsave a memory that I prefer morning meetings"
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.
ZGATRI Workspace
English | 简体中文
A self-hosted personal workspace for memory, ideas, projects, tools, and infrastructure. Built with vanilla HTML, CSS, JavaScript, and a Node.js/SQLite API, with an optional Hindsight memory backend.
Quick start · Screenshots · Workbench guide · Deployment
Features
Area | What the UI includes |
Projects | Cards, list view, status filters, project details and editing |
Memory | Categories, tags, timeline, related content and a detail panel |
Ideas | Card/list/board layouts, search, tags and quick capture |
Information collection | Sources, tasks, processing pipeline and logs |
Infrastructure | Servers, containers, domains, monitoring and security panels |
Tools | Shortcuts, categories, recent activity and script entries |
Security and settings | Credential metadata, members, devices, preferences and notifications |
Overview | A shared entry point for memory metrics and project cards |
Production defaults to actual data, refreshed every 30 seconds while visible. Projects, ideas, tools, memories, ingestion jobs, host resources, containers and the configured origin certificate use backend data. Local static preview explicitly uses synthetic demo content. External collection, member/device management and alerts remain marked as unconnected. See live backend coverage.
Related MCP server: everos-mcp
Quick start
Requires Node.js 22.18 or later. From the repository root:
git clone https://github.com/Ti08X/zgatri-workspace.git
cd zgatri-workspace
npm run preview:workbenchOpen http://127.0.0.1:8898/#projects. The preview serves static files only and does not connect to a production database. See the workbench guide for API setup, configuration and checks.
Screenshots
All images below are regenerated from the public code with sanitized, synthetic data. Addresses, accounts, device details and memory entries are examples. Text-bearing private thumbnails are replaced with synthetic artwork. Browse all 16 desktop/mobile images.
Projects

Memory

Ideas

Information collection

Infrastructure

Tools

Security and settings

Overview

Project structure
workbench/
portal/ Current workspace UI and assets
server.mjs Node.js/SQLite workspace API
scripts/ Static preview and isolated API/UI checks
docs/ Redesign notes and sanitized screenshotsThe other src/, server/, public/ and integration directories retain the earlier Memory Hub implementation. Its historical guide documents that separate implementation; it is not the entry point for the redesigned workspace.
Documentation
Keep authentication at the reverse proxy and the API on loopback. This source distribution contains no production database, private archive, credentials or backups.
License
This server cannot be deployed
Maintenance
Related MCP Connectors
- KogniteOAuthdev.kognite
Hosted agent memory: store, search, and recall facts across sessions from any MCP client.
Shared memory for connected AI tools. Projects, rules and skills over MCP. OAuth or API key.
- JustOnceOAuthai.justonce
Persistent memory for AI assistants — one shared, OAuth-secured vault for every MCP client.
Private-by-default, local-first memory/context/task orchestrator for MCP apps and agents.
Related MCP Servers
- FlicenseNot gradedqualityFmaintenanceEnables remote MCP clients to securely access AgentMemory tools like memory_recall, memory_smart_search, and memory_save through OAuth 2.1 and Streamable HTTP, while keeping the AgentMemory backend secret private.-
- FlicenseNot gradedqualityCmaintenanceEnables MCP clients to interact with an EverOS v2 memory store over local Streamable HTTP, providing read-only memory tools and an opt-in, approval-gated write capability.-
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to store, search, and link typed knowledge entities such as memories, people, projects, tasks, and documents over a Streamable HTTP MCP endpoint. Combines hybrid BM25 and vector search with a knowledge graph, and separates short-lived session context from durable memory that future agents can trust.10AGPL 3.0

brain-mcpofficial
AlicenseNot gradedqualityAmaintenanceGives MCP clients a small remote Streamable HTTP tool surface for a Brain memory database, letting agents store facts, retrieve current-truth answers with evidence, delete memories, and check their effective identity. Each call is scoped by a customer identifier for per-customer isolation, and the endpoint can optionally be secured as an OAuth 2.1 resource server.Apache 2.0