• 

  AI Emergency Assistant – Main Dashboard

• 

  Smoke Emergency Detection and Safety Response

• 

  High Heat Detection and Emergency Response

• 

  Security Alert for Unknown Person Detection

• 

  Voice-Based Emergency Input

• 

  Incident History and Recent Emergency Records

Inspiration

During an emergency, people need quick and understandable assistance. We wanted to build a simple system that could recognize common emergency situations and provide an immediate response through text or voice.

This inspired us to create AI Emergency Assistant, an MCP-powered safety assistant for smoke, high heat, and security alerts.

What it does

The application allows users to describe an emergency using text or voice.

It can identify:

• 🚬 Smoke emergencies
• 🔥 High-temperature emergencies
• 🚨 Security/intruder alerts

After analyzing the request, the system identifies the emergency type, assigns a severity level, and provides safety guidance.

How we built it

The project uses a web-based frontend with HTML, CSS, and JavaScript.

The backend is built with Python and Flask. An AI-agent layer processes the user's request and communicates with an MCP client.

The MCP client connects to our self-hosted MCP server using Streamable HTTP. The MCP server exposes dedicated tools for:

• detect_smoke
• detect_heat
• detect_security

The application also uses the browser's Web Speech API for voice input and LocalStorage for recent incident history.

Architecture

User ↓ Web Interface ↓ Flask Backend ↓ AI Agent ↓ MCP Client ↓ MCP Server ↓ Emergency Tools

What we learned

Through this project, we learned how MCP can be used to create modular

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for AI Emergency Assistant

Built With

• aiagent
• artificialintelligence
• css
• flask
• html
• javascript
• localstorage
• mcp
• python
• restapi
• streameableapi
• webspeechapi