ContextKeep by mordang7 - MCP Server

ContextKeep Banner

ContextKeep 🧠

Infinite Long-Term Memory for AI Agents

ContextKeep is a powerful, standalone memory server that gives your AI tools (Claude, Cursor, VS Code, Gemini, OpenCode) a persistent, searchable brain. Stop repeating yourself—let your AI remember.

Features • Installation • Usage • Web Dashboard • Configuration

🌟 Features

♾️ Infinite Context: Store unlimited project details, preferences, and snippets.
💰 Save Money & Tokens: Reduce context window usage by only retrieving relevant memories, lowering API costs.
🔌 Universal Compatibility: Works with any MCP-compliant client via Stdio (Local) or SSE (Remote).
🖥️ Modern Web Dashboard: Manage your memories visually with Grid, List, and Calendar views.
🔒 Privacy First: 100% local storage. Your data never leaves your machine.
🔎 Smart Search: Find exactly what you need with semantic and keyword search.
🐧 Linux Service: Runs silently in the background as a system service.

ContextKeep Showcase

🚀 Installation

Prerequisites

Python 3.10 or higher
Git (optional)

Quick Start

Clone the repository:

git clone https://github.com/mordang7/ContextKeep.git
cd ContextKeep

Run the Installer:
- Linux/Mac:
```
python3 install.py
```
- Windows:
```
python install.py
```
Follow the Wizard: The installer will create a virtual environment, install dependencies, and generate a custom configuration file for you.

🔌 Configuration

After installation, you will find a mcp_config.json file in the root directory. Copy its contents into your AI client's configuration.

Example Configurations

1. Claude Desktop / Gemini CLI (Local)

{
  "mcpServers": {
    "context-keep": {
      "command": "/absolute/path/to/ContextKeep/venv/bin/python",
      "args": ["/absolute/path/to/ContextKeep/server.py"]
    }
  }
}

2. Remote Access (SSH)

Perfect for running ContextKeep on a home server (e.g., Raspberry Pi) and accessing it from your laptop.

{
  "mcpServers": {
    "context-keep": {
      "command": "ssh",
      "args": [
        "-i", "/path/to/private_key",
        "user@192.168.1.X",
        "'/path/to/ContextKeep/venv/bin/python'",
        "'/path/to/ContextKeep/server.py'"
      ]
    }
  }
}

3. SSE Mode (Http)

If you prefer HTTP transport (great for OpenCode or web apps):

{
  "mcpServers": {
    "context-keep": {
      "transport": "sse",
      "url": "http://localhost:5100/sse"
    }
  }
}

🌐 Web Dashboard

ContextKeep includes a beautiful web interface to manage your memories.

URL: http://localhost:5000
Features:
- Dashboard: Overview of recent memories.
- Calendar: Visual timeline of your work.
- Search: Instant filtering.
- CRUD: Create, Read, Update, Delete memories manually.

To start it manually:

./venv/bin/python webui.py

🐧 Linux Service Setup (Optional)

Run ContextKeep as a background service (systemd) on Linux/WSL:

chmod +x install_services.sh
./install_services.sh

This will install:

contextkeep-server (Port 5100)
contextkeep-webui (Port 5000)

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

☕ Support the Project

If ContextKeep helps you build cool things, consider buying me a coffee!

_{Built with ❤️ by GeekJohn}