Microsoft Word MCP server

Semantic Workbench is a versatile tool designed to help prototype intelligent assistants quickly. It supports the creation of new assistants or integration of existing ones within a cohesive interface, providing a user-friendly UI for creating conversations, configuring settings, and exposing various behaviors. The workbench is framework-agnostic and connects to assistants via a RESTful API.

Installation

Using GitHub Codespaces (Recommended)

GitHub Codespaces provides a turn-key development environment without needing to configure your local machine:

1. Open the repository in GitHub
2. Click the "Code" button
3. Select the "Codespaces" tab
4. Click "Create codespace on main"

This creates a cloud-based development environment with all dependencies pre-installed.

Local Development Setup

If you prefer to install locally:

Prerequisites

• Docker (recommended for dev containers)
• VS Code
• Python 3.10 or higher
• Node.js 18 or higher

Using VS Code and Dev Containers

1. Clone the repository
2. Open the workspace file:

code semantic-workbench.code-workspace

3. When prompted to "Reopen in Container", click "Reopen in Container"
4. VS Code will build the development container with all dependencies

Manual Setup

If not using dev containers, follow these steps:

1. Install the workbench service dependencies:

cd workbench-service
pip install -e .

2. Install the workbench app dependencies:

cd workbench-app
npm install

Running the Workbench

Using VS Code

1. Open the VS Code Command Palette (Ctrl/Cmd+Shift+P)
2. Select "Run and Debug" (or press Ctrl/Cmd+Shift+D)
3. Choose "semantic-workbench" from the dropdown menu
4. Click the play button to start the application

This will start both the backend service and the frontend app.

From Command Line

Run the provided script:

# For Linux/macOS
./tools/run-workbench-chatbot.sh

# For Windows
.\tools\run-workbench-chatbot.ps1

Accessing the Workbench

Open your browser and navigate to:

https://localhost:4000

You may receive a security warning about the self-signed certificate. Click "Advanced" and "Proceed to localhost" to continue.

Starting an Assistant Service

The workbench requires at least one assistant service to interact with. Choose one of these sample assistants:

Echo Bot (No API Keys Required)

In VS Code:

1. Go to "Run and Debug" (Ctrl/Cmd+Shift+D)
2. Select "examples: python-01-echo-bot"
3. Click the play button

This starts a simple assistant that echoes your messages back.

Simple Chatbot (Requires LLM API Keys)

In VS Code:

1. Go to "Run and Debug" (Ctrl/Cmd+Shift+D)
2. Select "examples: python-02-simple-chatbot"
3. Click the play button

You'll need to provide your LLM API keys either in a .env file or through the assistant configuration page.

Using the Workbench

Creating a New Assistant

1. On the dashboard, click the "New Assistant" button
2. Select your assistant service from the dropdown menu
3. Provide a name for your assistant
4. Click "Save"

Starting a Conversation

1. On the dashboard, click the "New Conversation" button
2. Provide a title for the conversation
3. Click "Save"

Adding an Assistant to a Conversation

1. In the conversation window, click the conversation canvas icon
2. Select your assistant from the list
3. The assistant will be added to your conversation

Interacting with Your Assistant

1. Type your message in the text box at the bottom
2. Press Enter or click Send
3. The assistant will respond in the conversation window

Resetting the Workbench

If you need to reset all service data:

# For Linux/macOS
./tools/reset-service-data.sh

# For Windows
.\tools\reset-service-data.ps1

Alternatively, you can delete the data directory:

rm -rf ~/workbench-service/.data

Refreshing the Development Environment

To completely refresh your development environment:

# From repository root
make clean install

For a faster refresh of just the workbench components:

# From repository root
make clean

# Then from workbench-app directory
cd workbench-app
make install

# Then from workbench-service directory  
cd ../workbench-service
make install

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.