# Leave Management System using MCP (Model Context Protocol)

A **minimal, AI-ready Leave Management System** built using **FastMCP (Model Context Protocol)**.  
This project demonstrates how backend business logic can be exposed **directly to AI clients** (like Claude Desktop), without REST APIs or prompt hacks.

---

## ✨ Why This Project Exists

Most of us:
- Watch reels
- Save YouTube tutorials
- Bookmark blogs

…but never execute.

This project started when I searched:

> **“fast api backend for employee management system”**

and found this Medium article:  
👉 *Building a Leave Management System with FastAPI and FastMCP* by Madhuri

Instead of just reading it, I asked:

> **“Why not build it myself?”**

That curiosity turned into execution — and this repo is the result.

---

## 🧠 What This Project Demonstrates

- What **MCP (Model Context Protocol)** actually looks like in practice
- How AI models can **discover and call backend functions**
- How to build an **AI-first backend**, not just an API
- How Claude Desktop interacts with real application logic

---

## ⚙️ Features

### MCP Tools
- `get_leave_balance(employee_id)`
- `apply_leave(employee_id, leave_dates)`
- `get_leave_history(employee_id)`

### MCP Resource
- `greeting://{name}`

### System Details
- In-memory mock database
- Claude Desktop compatible (stdio-based MCP server)
- No REST APIs, no frontend — pure MCP

---

## 🏗️ Project Structure

leave-manager-mcp/
├── main.py # MCP server (Claude-compatible)
├── requirements.txt # Dependencies
└── README.md

yaml
Copy code

---

## ▶️ How It Works (Big Picture)

Claude Desktop
↓
Model Context Protocol (MCP)
↓
FastMCP Server (Python)
↓
Business Logic (Leave Management)

yaml
Copy code

Claude **discovers tools**, understands their schema, and calls them directly.

---

## 🚀 Getting Started

### 1️⃣ Create & activate virtual environment
```bash
python -m venv .venv
source .venv/bin/activate     # Windows: .venv\Scripts\activate
2️⃣ Install dependencies
bash
Copy code
pip install -r requirements.txt
3️⃣ Run MCP server (manual testing)
bash
Copy code
python main.py
⚠️ For Claude Desktop, Claude itself starts this process via MCP configuration.

🤖 Claude Desktop Configuration (Windows Example)
json
Copy code
{
  "mcpServers": {
    "LeaveManagerPython": {
      "command": "C:\\Users\\<USERNAME>\\path\\to\\.venv\\Scripts\\python.exe",
      "args": [
        "C:\\Users\\<USERNAME>\\path\\to\\main.py"
      ],
      "cwd": "C:\\Users\\<USERNAME>\\path\\to\\leave-manager-mcp"
    }
  }
}
🔁 Restart Claude Desktop after updating the config.

🧪 Example Prompts (Claude)
rust
Copy code
Check leave balance for employee E001
Apply leave for employee E002 on 2025-04-17 and 2025-04-18
Show leave history of E001
Use greeting://Akash
Claude will automatically call the correct MCP tools.

⚠️ Key Learnings
MCP requires a pure stdio process

FastAPI & MCP must be separated

Claude Desktop caches MCP configs aggressively

Execution teaches more than tutorials

🔮 Future Improvements
Add FastAPI (dual-mode: REST + MCP)

Persist data using PostgreSQL

Add authentication & roles

Build AI agents on top of MCP tools

Production deployment

📌 Final Thought
Reading tutorials gives comfort.
Watching reels gives motivation.

But execution gives clarity.

This repo exists because of curiosity — not perfection.

yaml
Copy code

---

## ✅ Why This README Works

✔ Clean headings  
✔ Clear story hook  
✔ Explains MCP simply  
✔ Recruiter / dev friendly  
✔ Medium-article ready  
✔ No fluff, no noise  

---

### 🚀 Next (Recommended)
Now that GitHub is ready, next we should:
1. **Write the Medium article using this repo**
2. **Add FastAPI back cleanly (dual-mode)**
3. **Create an AI agent on top of this**

Just say **which one** and we continue 🔥