Update README.md

README.md

@@ -1,13 +1,63 @@
 ---
-title: Ipmentor
-emoji: 📊
-colorFrom: yellow
-colorTo: indigo
+title: IPMentor
+emoji: 🌐
+colorFrom: blue
+colorTo: green
 sdk: gradio
-sdk_version: 5.32.1
+sdk_version: 5.33.1
 app_file: app.py
 pinned: false
 license: mit
+short_description: IPv4 networking toolkit with verified calculations
+tags:
+  - Agents-MCP-Hackathon
+  - mcp-server-track
+  - networking
+  - education
+  - ipv4
+  - subnet-calculator
+  - ai-tutoring
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# IPMentor 🌐
+
+**IPMentor** is an IPv4 networking toolkit designed as verified computational tools for AI tutoring systems. Built for the **Gradio MCP Hackathon 2025**, this project demonstrates how MCP can bridge AI tutoring systems with specialized computational tools, creating more reliable and cost-effective educational experiences.
+
+## 🎯 Hackathon Track: MCP Server/Tool
+
+This Gradio app serves as both an interactive web interface and an **MCP Server**, providing three core networking tools that AI agents can access through the Model Context Protocol:
+
+- `ip_info` - Analyze IPv4 addresses and subnet masks
+- `subnet_calculator` - Perform subnet calculations with multiple division methods  
+- `generate_diagram` - Create visual network diagrams
+
+## 🏆 Competing for Mistral AI Choice Award
+
+This project uses **Mistral Small 3.1 24B Instruct** in the AI chatbot demo, showcasing how smaller, efficient models can handle educational interactions while delegating precise calculations to IPMentor's verified tools.
+
+## 🎥 Demo Video
+
+Video demonstration: [assets/ipmentor-demo.mp4](assets/ipmentor-demo.mp4)
+
+## 🤖 Live AI Chatbot Demo
+
+Experience IPMentor in action with an Mistral Small 3.1 24B Instruct: [ipmentor-demo](https://huggingface.co/spaces/Agents-MCP-Hackathon/ipmentor-demo)
+
+## 💡 Why IPMentor?
+
+Current AI tutoring faces a challenge: while LLMs can perform calculations, they occasionally make errors and using powerful models for every calculation is expensive. IPMentor solves this by:
+
+- **Verified Calculations**: All subnet mathematics uses dedicated algorithms, eliminating computational errors
+- **Cost-Effective AI**: Smaller models handle pedagogy while IPMentor handles precise calculations
+- **Educational Focus**: Designed specifically for networking education scenarios
+
+## 🔗 Links
+
+- **GitHub Repository**: [https://github.com/DavidLMS/ipmentor](https://github.com/DavidLMS/ipmentor)
+- **AI Chatbot Demo**: [https://huggingface.co/spaces/Agents-MCP-Hackathon/ipmentor-demo](https://huggingface.co/spaces/Agents-MCP-Hackathon/ipmentor-demo)
+
+## 🛠️ Technical Architecture
+
+Built with Python, Gradio, native IPv4 algorithms, D2 for diagrams, MCP protocol support, and Pydantic validation. This creates a reliable foundation for AI-powered networking education.
+
+**Integration Focus**: IPMentor is designed to complement [LearnMCP-xAPI](https://github.com/DavidLMS/learnmcp-xapi) for comprehensive AI tutoring systems. While IPMentor provides verified computational tools, LearnMCP-xAPI maintains persistent learning records, enabling AI tutors that can both perform accurate calculations and adapt to individual student learning patterns over time.