AI Personal Hub

Provides tools to access local notes, GitHub, YouTube, Gmail, Steam, and summarization via MCP.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "harsh-1807-ai-personal-hub": {
      "command": "python",
      "args": [
        "-m",
        "mcp_server.server"
      ],
      "env": {
        "STEAM_ID": "YOUR_STEAM_ID",
        "GITHUB_TOKEN": "ghp_XXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
        "STEAM_API_KEY": "YOUR_STEAM_API_KEY"
      }
    }
  }
}

AI Personal Hub MCP Server exposes a set of practical tools that let you access and manipulate local data, GitHub content, YouTube data, Gmail messages, Steam playtime, and text summarization through a unified MCP interface. It enables your LM assistant to call specific MCP tools to fetch, list, or modify information across your connected sources, all through a clean, extensible server built to work with LM Studio and Flask UI.

How to use

You interact with the MCP server by issuing prompts to an MCP-enabled client (for example, LM Studio) that is configured to connect to this server. Start the MCP server, then connect your client to the provided HTTP or STDIO endpoint. When you ask for information or actions such as listing local notes, fetching a GitHub file, reading your latest emails, or summarizing recent messages, your client will route the request to the appropriate MCP tool. The server responds with the requested data and, when appropriate, links or content ready for display in your UI.

Practical usage patterns include: creating a local note inventory, listing files in a GitHub repo, fetching specific file contents, listing YouTube liked items, reading recent Gmail messages, identifying Steam playtime, and generating a concise summary of recent data. You can mix and match tools in a single prompt, and the system will call the necessary tools to fulfill your request.

How to install

Prerequisites you need before installing: Python, a virtual environment tool, and access to LM Studio for running a local OpenAI-compatible server. You should also have or create the required credentials and tokens for each external service you plan to use (GitHub, YouTube, Gmail, Steam).

python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt

# Start LM Studio locally (recommended): ensure it runs on http://localhost:1234
# Then register the MCP server as shown below and start the MCP server

# Start the MCP server
python -m mcp_server.server

Configuration and running notes

The MCP server runs as a separate process from the Flask UI. You will typically run the UI and the MCP server in separate terminals. The server is designed to be started with Python and the module path to the MCP server entry point.

Available tools

list_local_files

List available .txt files in the notes/ directory.

fetch_local_file

Fetch the content of a local file by path, enabling reading inside prompts.

github_repos

List GitHub repositories you have access to or specified by user.

github_commits

Fetch commit data for a given repository.

github_list_files

List files in a specified GitHub repository path.

github_file_content

Retrieve the content of a file in a GitHub repository.

github_issues

List issues for a GitHub repository.

github_issue

Fetch a specific issue from a GitHub repository.

yt_liked_videos

List a sample of your liked YouTube videos (LL).

ytm_liked_songs

List a sample of your liked songs on YouTube Music (LM).

yt_playlist

Fetch contents of a YouTube playlist.

read_emails

Read recent emails from Gmail (OAuth-based access).

steam_games

Show recently owned Steam games and playtime for your account.

summarize

Run a summarization prompt against your current context.