Knowledge Base MCP Server

This project implements a Model Context Protocol (MCP) server that provides tools for managing a knowledge base through API interactions. It allows AI assistants to create, update, and search notes in your knowledge base.

Overview

The Knowledge Base MCP Server provides the following functionality:

• Add new notes with titles, content, and tags
• Update existing notes
• Find note IDs by searching for keywords in titles or content

This server acts as a bridge between AI assistants and your knowledge base API, enabling structured interaction with your notes database.

Prerequisites

• Node.js (v14 or higher recommended)
• npm or yarn
• A running API server at http://localhost:3000 that provides note management functionality

Installation

1. Clone this repository
2. Install dependencies:

   npm install

   or

   yarn install

Dependencies

This project relies on the following packages:

• @modelcontextprotocol/sdk - The Model Context Protocol SDK
• axios - For making HTTP requests to the knowledge base API
• zod - For schema validation

Building the Server

Build the server with

npm run build

Running the Server

Start the server with:

node index.js

The server will run on stdio, allowing it to communicate with clients that implement the MCP client protocol.

Testing Your Server To test your MCP server with Claude for Desktop:

Follow the guide from Anthropic at: Testing Your Server with Claude for Desktop

This guide will walk you through the process of connecting your MCP server to Claude and verifying that the tools are working correctly.

Available Tools

1. add_note

Creates a new note in the knowledge base.

Parameters:

• title (string): Title of the note
• content (string): Content of the note
• tags (array): Array of tag objects with id, name, and color properties

2. update_note

Updates an existing note in the knowledge base.

Parameters:

• id (string): The ID of the note to update
• title (string, optional): New title for the note
• content (string, optional): New content for the note
• tags (array, optional): New tags for the note

3. find_note_id_by_keyword

Searches for notes containing a specific keyword in their title or content.

Parameters:

• keyword (string): A keyword to search for in note titles or content

API Integration

This server expects a knowledge base API running at http://localhost:3000 with the following endpoints:

• POST /api/notes - Create a new note
• PUT /api/notes/:id - Update an existing note
• GET /api/notes - Get all notes

Note Structure

A note in the knowledge base has the following structure:

interface Note {
  id: string;
  title: string;
  content: string;
  tags: Array<{
    id: string;
    name: string;
    color: string;
    notesCount?: number;
  }>;
  updatedAt: string;
}

Error Handling

The server includes error handling for API requests. If an error occurs, it will be logged to the console and returned in the tool response.

License

[Your License Here]

Contributing

[Your Contribution Guidelines Here]