Sponsored by Deepsite.site

Personal Context Technology MCP Server

Created By
mikhashev8 months ago
Personal Context Technology for AI Personalization MCP server
Content

Personal Context Technology MCP Server

This repository implements a server for Personal Context Technology (PCT) using the Model Context Protocol (MCP). It enables AI assistants like Claude to access and update your personalized context data, creating persistent memory between sessions.

What is Personal Context Technology?

Personal Context Technology allows you to structure and manage personal data that AI assistants can use to provide more personalized and relevant responses. With PCT:

  • Your context is preserved between AI sessions
  • You control where your data is stored (locally, cloud, etc.)
  • You define explicit instructions for how AI systems should use your data
  • You can update context as your preferences and situation change

Features

  • Persistent Context Storage: Store your preferences, goals, learning style, and other personal information
  • Privacy Controls: Instruction block defines which information is private/public
  • Section-based Access: Access full context or individual sections
  • Context Updates: Update specific fields via the updateContext tool
  • Version Tracking: Maintain history of context changes

Demo

Watch the Personal Context Technology MCP Server in action:

PCT MCP Server Demo

The demo shows how to:

  • Access personal context data using the MCP tool
  • Update context information using the updateContext tool
  • See how Claude's responses improve with personalized context

Installation

Prerequisites

  • Node.js (v16+)
  • npm (included with Node.js)

Setup Instructions

  1. Clone the repository
git clone https://github.com/mikhashev/pct-mcp-server.git
cd pct-mcp-server
  1. Install dependencies
npm install
  1. Build the server
npm run build
  1. Start the server
npm start

Connecting to Claude Desktop

To use this server with Claude Desktop:

Windows

  1. Open Claude Desktop
  2. Navigate to Settings > Developer > Edit Config
  3. This will open %AppData%\Claude Desktop\ directory
  4. Create or edit claude_desktop_config.json with the following content:
{
  "mcpServers": {
    "personal-context-server": {
      "command": "node",
      "args": ["PATH_TO_YOUR_PROJECT/pct-mcp-server/dist/index.js"],
      "env": {}
    }
  }
}

Replace PATH_TO_YOUR_PROJECT with the actual path to your project.

macOS

  1. Open Claude Desktop
  2. Navigate to Settings > Developer > Edit Config
  3. This will open ~/Library/Application Support/Claude Desktop/ directory
  4. Create or edit claude_desktop_config.json with the following content:
{
  "mcpServers": {
    "personal-context-server": {
      "command": "node",
      "args": ["/path/to/your/project/pct-mcp-server/dist/index.js"],
      "env": {}
    }
  }
}

Linux

  1. Open Claude Desktop
  2. Navigate to Settings > Developer > Edit Config
  3. This will open ~/.config/Claude Desktop/ directory
  4. Create or edit claude_desktop_config.json with the following content:
{
  "mcpServers": {
    "personal-context-server": {
      "command": "node",
      "args": ["/path/to/your/project/pct-mcp-server/dist/index.js"],
      "env": {}
    }
  }
}

Using the Personal Context Server

With Claude Desktop

  1. Start your PCT MCP server using npm start
  2. Restart Claude Desktop (if needed)
  3. In Claude, click the MCP tool icon (wrench/tool icon) in the input field
  4. You'll see "personal-context-server" with available tools and resources:
    • Resources: Access your context data (full context or by section)
    • Tools: Update specific fields in your context

Accessing Context Data

Claude can access your personal context data in two ways:

  1. Full Context: Select "All Sections" to give Claude access to your entire personal context
  2. Section-based: Select specific sections (basic_info, preferences, etc.) to share only certain parts of your context

Example prompt: "Using my context data, recommend some learning resources that would match my learning style."

Updating Context Data

You can update your personal context using the updateContext tool:

Example prompt: "Update my learning_style in preferences to 'visual and hands-on' because I've found I learn better with practical examples."

Customizing Your Context

The default personal context includes:

  • basic_info: Name, location, etc.
  • preferences: Communication style, learning preferences, etc.
  • instruction: Rules for how AI should use and update your context
  • metadata: Version tracking, update history

You can modify the default structure by editing the DEFAULT_CONTEXT object in src/storage/contextStorage.ts. Or just replace update file data [PATH_TO_YOUR_PROJECT/pct-mcp-server/data/personal_context.json] from template data in repo (https://github.com/mikhashev/personal-context-manager/blob/main/use-cases/self-education/personal_context_self_education_template.json) or another one.

Project Structure

pct-mcp-server/
├── .git/                      # Git repository data
├── .gitignore                 # Git ignore rules
├── LICENSE                    # License file
├── README.md                  # Project documentation
├── package-lock.json          # Lock file for npm dependencies 
├── package.json               # Project configuration and dependencies
├── tsconfig.json              # TypeScript configuration
└── src/                       # Source code
    ├── index.ts               # Main application entry point
    ├── resources/             # MCP resources implementation
    │   └── personalContext.ts # Personal context resource
    ├── storage/               # Storage implementation
    │   └── contextStorage.ts  # Context storage functionality
    ├── tools/                 # MCP tools implementation
    │   ├── contextSuggestion.ts  # Context suggestion tool
    │   └── updateContext.ts   # Update context tool
    └── utils/                 # Utility functions
        └── instructionHandler.ts  # Instruction handling logic

# Generated directories (not in repository)
# ----------------------------------------
# dist/                        # Compiled JavaScript files (gitignored)
# node_modules/                # Dependencies (gitignored)
# data/                        # User data (gitignored or contains only samples)
#   └── personal_context.json  # Personal context data (sample only) (don't change filename personal_context.json, update data if need)

Advanced Configuration

Using HTTP Transport

The default implementation uses stdio transport for development. For production use, you can switch to HTTP transport:

  1. Edit src/index.ts
  2. Uncomment the HTTP transport section
  3. Configure your desired port and CORS settings
  4. Rebuild and restart the server

! I reccomend use stdio transport for development locally on your machine always.

Multi-User Support

To support multiple users, modify the contextStorage.js file to store contexts in user-specific files based on user IDs.

Contributing

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

License

This project is released under the MIT License - See LICENSE that covers the project.

Acknowledgments

This project is based on the Personal Context Technology concept and implements the Model Context Protocol (MCP) developed by Anthropic to enable AI systems to access external context.

Recommend Servers
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
TimeA Model Context Protocol server that provides time and timezone conversion capabilities. This server enables LLMs to get current time information and perform timezone conversions using IANA timezone names, with automatic system timezone detection.
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
Serper MCP ServerA Serper MCP Server
Baidu Map百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Amap Maps高德地图官方 MCP Server
Context7Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
Playwright McpPlaywright MCP server
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
Tavily Mcp
WindsurfThe new purpose-built IDE to harness magic
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
ChatWiseThe second fastest AI chatbot™
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
DeepChatYour AI Partner on Desktop
CursorThe AI Code Editor
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code