Inspiration

The inspiration came from the critical need to protect strategic military flight corridors and high-value official delegations. We wanted to build a sovereign, zero-latency aerospace defense system that removes human delay during fast radar anomalies.

What it does

NEXUS is an AI-driven mission control center for airspace protection. It tracks live flight data like hypersonic propulsion speed, reactor core stability, and radar status. When an intruder is detected, the Gemini agent uses Model Context Protocol (MCP) tools to immediately log advanced variables—such as unique radar target IDs and live GPS coordinates inside the Romanian corridor—directly into a MongoDB Atlas cloud cluster, while automatically deploying the Sfinx-01 escort vehicle.

How we built it

The futuristic frontend panel is built with Python and Flask, hosted live on PythonAnywhere. The core logic runs on Google Cloud Agent Builder using Gemini 1.5 Pro. We utilized the Model Context Protocol (MCP) to bridge Gemini's advanced reasoning directly with a secure MongoDB Atlas database for real-time secure logging.

Challenges we ran into

Setting up the secure data pipeline and managing remote cloud processes on a lightweight environment was a challenge. We also had to configure precise database network filters to allow secure, uninterrupted real-time tracking from our backend.

Accomplishments that we're proud of

We are proud of building a fully functional, highly detailed cyber-defense ecosystem that bridges advanced AI reasoning with an active cloud database under 5 days, achieving a unique visual and technical harmony.

What we learned

We learned how to implement the Model Context Protocol (MCP) to expand an LLM's capabilities into database environments, moving far beyond a simple text chatbot into a real-world task agent.

Built With

Share this project:

Updates

posted an update

Project Update: Added Global Airspace Defense Simulation Dataset!I have successfully updated the GitHub repository with a massive strategic scenario dataset: global_defense_simulation.json.This new simulation demonstrates how the NEXUS system scales during extreme, high-density aerospace crises. It simulates the automated deployment of all 10,000 Sfinx-01 Omni-Fire interceptors managing simultaneous target tracking and threat neutralization against complex global air defense vectors.Crucially, this dataset showcases the core power of the Model Context Protocol (MCP). It proves that Gemini 1.5 Pro can reason through heavy telemetry loops and securely synchronize multi-target variables into MongoDB Atlas with zero latency—completely hands-free and without human delay. Check out the repository to inspect the active database logging structures!

Log in or sign up for Devpost to join the conversation.

posted an update

Project Update: Why a Zero-UI Autonomous Dashboard?

Unlike traditional applications that rely on manual buttons, NEXUS is engineered as a true autonomous agent. In a high-stakes airspace emergency, human reaction time is a vulnerability.

Our interface functions as a real-time tactical streaming terminal. The Gemini 1.5 Pro brain continuously scans the telemetry feed, detects anomalies, and executes database logic via MCP entirely on its own. The system eliminates human latency, proving that the best user interface for a split-second defense agent is one that operates completely hands-free.

Log in or sign up for Devpost to join the conversation.

posted an update

Project Update: Added Live Simulation & Telemetry Data!

I have just updated the GitHub repository with a realistic system telemetry dataset.

This update provides a clear view of the structured variables that our Flask backend processes and logs into MongoDB Atlas via the Model Context Protocol (MCP) whenever Gemini 1.5 Pro detects a critical airspace threat. Check out the updated README to see the exact JSON structure of a Sfinx-01 intercept mission!

Log in or sign up for Devpost to join the conversation.