mcp -on-raspi

This MCP server provides a simple note storage system with a custom note:// URI scheme for accessing individual notes. Each note has a name, a short description, and a text/plain payload. The server exposes a single prompt, summarize-notes, which can generate a summary of all stored notes. You can optionally pass a style argument (brief or detailed) to control the level of detail in the summary. The server also includes a tool called add-note, which lets clients add new notes by providing a name and the content. When a note is added, the server updates its internal state and notifies connected clients of the change in resources. Tools and prompts are designed to be combined during a session to manage and summarize notes efficiently.

Prerequisites:

• Python 3.11+ (or the Python version required by the MCP server project)
• Node.js and npm are not required for this Python-based MCP server
• The MCP tooling (uv/uvx) installed on your system (see uv CLI installation below)

Installation steps:

1. Install the MCP tooling (uv/uvx) on your system. For example, using pipx:

python3 -m pip install --user pipx
python3 -m pipx ensurepath
pipx install uv

2. Clone or download the MCP server repository (this project):

git clone https://github.com/daikw/mcp-server-on-raspi.git
cd mcp-server-on-raspi

3. If you want to run it locally in development mode, you can start the server using uv/uvx as described in the Quickstart. For a published setup, ensure the server is installed and available in your PATH as shown below.
4. Run the server in published mode (example):

uvx mcp-server-on-raspi

5. If you prefer a development run with explicit directory context, you can use the alternative command shown in the README’s development example:

uv --directory /path/to/mcp-server-on-raspi run mcp-server-on-raspi

Notes:

• Ensure you are in an environment where the MCP tooling (uv/uvx) is accessible from your PATH.
• If you plan to publish to PyPI or distribute, follow the project’s standard packaging steps using uv publish (see Development section in the README).

Tips and common issues:

• The server uses a lightweight note storage model. If you encounter issues with notes not appearing, verify the server state file or in-memory store used by the implementation.
• When debugging, the MCP Inspector is recommended. It’s designed to connect to MCP servers over stdio and provide a debugging UI.
• If you’re running in a container or restricted environment, ensure the uv/uvx binary can access stdio and that you provide any required environment variables for authentication or configuration.
• For development deployments, you can modify the README’s example configurations to point to your local directory or to a packaged build as needed.
• The environment variable guidance in the README’s publishing section (UV_PUBLISH_TOKEN, UV_PUBLISH_USERNAME, UV_PUBLISH_PASSWORD) is only required if you are publishing to PyPI via UV tooling.