# 🖥️ PC Health Checker Assistant using MCP

An AI-powered PC Health Assistant built using the **Model Context Protocol (MCP)**, **LangChain**, and **Google Gemini**. Developed a MCP server from scratch using the official Python MCP SDK to enable AI-agent interaction with local system resources, demonstrating the complete MCP client-server architecture.

---

## Quick Preview
<img width="1632" height="722" alt="Screenshot from 2026-07-21 16-05-14" src="https://github.com/user-attachments/assets/f2f36ccc-2c2a-43d4-a8b9-77d53d301645" />

##  Project Objective

The objective of this project is to demonstrate how an AI agent can communicate with a custom-built MCP Server to access operating system information through standardized tools.

This project serves as a beginner-friendly implementation of:

- Model Context Protocol (MCP)
- MCP Server
- MCP Client
- LangChain Agents
- Gemini Tool Calling
- AI + System Integration

---
## Technology Stack

| Category | Technology |
|----------|------------|
| Programming Language | Python 3.12 |
| AI Framework | LangChain |
| LLM | Google Gemini 3,1 Flash Lite |
| Protocol | Model Context Protocol (MCP) |
| MCP SDK | Official MCP Python SDK |
| System Monitoring | psutil |
| OS Information | platform |
| Environment Variables | python-dotenv |

---

##  Features

- Check CPU Usage
- Check Memory Usage
- Check Disk Usage
- Retrieve Operating System Information
- Natural Language Interaction
- Custom MCP Server built from scratch
- LangChain Agent integrated with Gemini
- Modular architecture



---

## Project Architecture

```
                         User
                           │
                           ▼
                 Natural Language Query
                           │
                           ▼
                      main.py
                           │
                           ▼
                LangChain Agent
                           │
                           ▼
                  Google Gemini
                           │
          Chooses Appropriate Tool
                           │
                           ▼
               LangChain Tool Wrapper
                           │
                           ▼
                     MCP Client
                           │
═══════════════════════════════════════
          Model Context Protocol
═══════════════════════════════════════
                           │
                           ▼
                     MCP Server
                           │
                           ▼
                  System Tools
                           │
                           ▼
              psutil / platform
                           │
                           ▼
                  Operating System
```





---



## Example Conversation


### User

```
Show me the operating system information.
```

### Assistant

```
Operating System : Linux

OS Version : 7.0.0-28-generic

Architecture : x86_64

CPU Cores : 2

Logical Processors : 4

Python Version : 3.12.3
```

---

## Learning Outcomes

By completing this project I have understand:

- What is Model Context Protocol (MCP)
- How MCP Server works
- How MCP Client communicates with an MCP Server
- How LangChain integrates with MCP
- How Gemini performs tool selection
- Building AI agents with external tools
- Designing modular AI architectures

---

##  Installation

### Clone the Repository

```bash
git clone https://github.com/Parisa-Reza/PC-health-checking-MCP.git

cd PC-health-MCP
```

---

### Create Virtual Environment

Linux

```bash
python3 -m venv venv

source venv/bin/activate
```

---

### Install Dependencies

```bash
pip install -r requirements.txt
```

---

### Create Environment File

Create a file named

```
.env
```

Add your Gemini API key.



---

### Running the Project

Run the application.

```bash
python main.py
```

Example:

```
You:

How much CPU am I using?
```

---




## License

This project is intended for educational and learning purposes.