Welcome Text Generator

You can automatically generate professional onboarding texts for new employees by running a dedicated MCP server. It analyzes free-form employee information, extracts structured data, and assembles modular, customizable welcome messages using a five-module template system. This helps you scale onboarding while keeping messages consistent and personalized.

How to use

To create onboarding messages, you connect your MCP client to the Welcome Text Generator MCP Server. You first extract the employee data from a free-form description, then generate a modular welcome text using the predefined templates. You can progressively compose the final message by selecting which modules to include and choosing intro and closing variants.

How to install

Prerequisites you need before installing: Node.js version 18 or higher, and the Claude Desktop App for configuring the MCP client.

Step 1. Install the MCP server globally using npm.

npm install -g welcome-text-generator-mcp

Alternative for local development: clone the repository, install dependencies, and build.

git clone https://github.com/goodfel10w/WelcomeTextGenerator.git
cd WelcomeTextGenerator
npm install
npm run build

Step 2. Configure Claude Desktop to run the MCP server. Add a server entry with the runtime command and path to the built bundle.

{
  "mcpServers": {
    "welcome-text-generator": {
      "command": "node",
      "args": [
        "C:\\Pfad\\zum\\Projekt\\dist\\index.js"
      ],
      "env": {}
    }
  }
}

If you installed via npm for global use, Claude Desktop can run the package directly. Add a server entry that uses npx to invoke the MCP server.

{
  "mcpServers": {
    "welcome-text-generator": {
      "command": "npx",
      "args": ["welcome-text-generator-mcp"],
      "env": {}
    }
  }
}

Step 3. Restart Claude Desktop to load the MCP server.

Additional content and configuration

The MCP server supports a modular template system with five modules and a data storage mechanism. You can store extracted employee data and list all saved records. The server runs locally via Node.js and is initiated by the commands shown in the installation steps.

During setup, you may manage the MCP server configuration in JSON form within Claude Desktop. The two example configurations above demonstrate both a direct Node invocation and a local npm-based approach. Ensure the chosen method aligns with your environment and security practices.

Module overview and usage notes

The server modules cover greeting, competencies, optional fun facts, career history, and a closing message. You can enable or disable modules and select introduction and closing variants to tailor each onboarding message.

All textual content uses placeholders like [NAME] to insert the actual employee name at generation time. When you customize messages, you will combine the selected modules into a coherent, personalized welcome note.

Data storage and management

Extracted employee data is automatically stored for future use. The data is kept in your project’s data directory, with entries containing an identifier, timestamp, source, and the extracted data payload. Use the built-in tools to list or review stored records as needed.

Troubleshooting and tips

If Claude Desktop cannot load the MCP server after configuration, verify that the command path to the script is correct and that Node.js is installed and accessible in your system PATH. Check for typos in the JSON snippets and ensure there is a valid working directory for the MCP server. For data storage issues, confirm write permissions to the data folder and that the server has the necessary access rights to create and update files.

Available tools

extract_from_text

Extracts structured employee data from free-form text using defined parsing rules.

generate_modular_welcome_text

Generates a modular onboarding message based on the extracted data and selected template options.

list_extracted_data

Lists all stored employee data entries from the server storage.