employee-onboarding-mcp by sanjbhat - MCP Server

Employee Onboarding MCP Server

An Model Context Protocol (MCP) server that guides new employees through the onboarding process using AI assistants. This server provides interactive (via Cline in VS Code), step-by-step onboarding workflows with progress tracking and configurable content.

This is currently modeled for Dialtone, but can be adopted to any team.

🛠️ Features

Interactive Onboarding: AI-guided step-by-step process for new employees
Progress Tracking: Persistent storage of employee progress across sessions (locally), with scope to central hosting and tracking in the future.
Configurable Workflows: HTML-based configuration files for each onbaording step.
Auto-Detection: Automatic employee identification using corporate credentials

🚀 Getting Started

Prerequisites

Signup for GitHub Copilot aka.ms/copilot
VS Code with Cline extension aka.ms/cline

Installation

Clone or download this repository
Run the setup script:
```
.\scripts\setup.ps1
```
Restart VSCode
Start chatting in Cline: "Help me start my onboarding process"

That's it! The setup script handles everything automatically.

💬 Usage Examples

For New Employees

You: "I'm starting my onboarding today"
AI: "Welcome! Let me initialize your onboarding process..."

You: "What's my current step?"
AI: "You're on Step 2: SAW Device Request. Here's what you need to do..."

You: "Help me request a SAW device"
AI: "I'll help you submit a SAW device request. What type of device do you need?"

For Onboarding Buddies

You: "Show me all employees I'm mentoring"
AI: "Here are your mentees and their progress..."

You: "How is John Doe doing with onboarding?"
AI: "John is on Step 3 with 67% completion. He completed account setup and SAW device request..."

📁 Configuration

HTML Step Configuration

Steps are configured using markdown/html files in config/steps/: For Onboarding Buddies: This folder is where you can add your team specific on-boarding steps in html format.

Edit HTML/Markdown files in config/steps/ to customize onboarding steps
Update resource files in config/resources/ to add new links/videos or even use a url.

<div class="onboarding-step" data-step-id="1" data-step-type="account_setup">
  <h2>Account Setup</h2>
  <p>Set up your basic company accounts...</p>
  
  <div class="instructions">
    <h3>What You Need to Do:</h3>
    <ol>
      <li>Visit <a href="https://portal.company.com">Company Portal</a></li>
      <li>Complete security setup...</li>
    </ol>
  </div>
  
  <div class="completion-criteria">
    <h3>This step is complete when:</h3>
    <ul>
      <li>✅ You can log into the company portal</li>
    </ul>
  </div>
</div>

🏗️ Architecture

employee-onboarding-mcp/
├── src/                          # TypeScript source code
│   ├── index.ts                  # Main MCP server
│   └── utils/
│       ├── employee-identifier.ts # Employee detection & profiles
│       └── config-parser.ts      # HTML configuration parser
├── config/                       # HTML configuration files
│   ├── steps/                    # Onboarding step definitions
│   ├── resources/                # Wiki links, videos, etc.
├── data/                         # Employee progress data
│   ├── employees/                # Individual progress files
│   └── backups/                  # Automated backups
└── scripts/                      # Setup and utility scripts

This Employee Onboarding MCP server is built on the Model Context Protocol (MCP) standard, which enables AI assistants to communicate with external tools and data sources through a standardized JSON-RPC interface. The server runs as a local Node.js process on your machine, storing employee data in local JSON files while exposing 9 specialized tools (like start_onboarding, complete_step) that Cline can invoke through the MCP SDK.

🔧 Development

Building

npm install
npm run build

Manual Configuration

If you prefer manual setup, add to your MCP settings:

Cline (VS Code):

{
  "mcpServers": {
    "employee-onboarding": {
      "command": "node",
      "args": ["C:/path/to/employee-onboarding-mcp/build/index.js"],
      "env": {
        "ONBOARDING_DATA_PATH": "C:/path/to/data",
        "ONBOARDING_CONFIG_PATH": "C:/path/to/config"
      }
    }
  }
}

📊 Data Storage

Employee profiles: JSON files in data/employees/
Progress tracking: Automatic timestamps and completion status
Backups: Daily automated backups to data/backups/
Privacy: Data stored locally, not transmitted externally

🔍 Troubleshooting

Common Issues

"Employee registration required"

Run: register_employee tool with your email and name

TypeScript compilation errors

Ignore them - they don't affect functionality
Or install dependencies: npm install

MCP server not connecting

Check that Node.js is installed
Verify the build path in MCP settings
Restart your AI assistant

No onboarding steps found

Ensure config/steps/ contains HTML files
Check file naming: step-1-*.html, step-2-*.html, etc.

Happy Onboarding! 🎉

📜 License

MIT License - Feel free to customize for your organization's needs.