Cyber Trap AI

💡 Inspiration

The internet is full of automated bots and malicious scanners looking for vulnerabilities. Instead of just building a standard firewall, I wanted to build an active defense system. My inspiration was to create a "digital spiderweb"—a honeypot that traps attackers by pretending to be vulnerable, allowing us to study their behavior safely.

⚙️ What it does

Cyber Trap AI sets up decoy endpoints (like /admin, /login, /.env) that look like real vulnerable web pages to hackers. When automated bots or attackers try to access these pages, the system immediately captures their IP addresses, device types, timestamps, and request payloads. It then visualizes this data on a real-time analytics dashboard while securing the logs using advanced cryptography.

🛠️ How we built it

The system is built using pure Python 3 standard libraries to ensure it can run on any system (even older machines or lightweight servers) without heavy external dependencies. It includes:

1. The Trap Server: Uses Python's http.server to catch requests.
2. The Aegis Logger: Secures every log entry using SHA-256 Hash Chaining, making the logs tamper-evident.
3. The Frontend: HTML/Tailwind CSS dashboard for real-time visualization.

🚧 Challenges we ran into

The biggest challenge was ensuring data integrity without using heavy external databases like PostgreSQL. I solved this by implementing an atomic, lightweight file-write mechanism and local cryptographic chaining. Another challenge was running it efficiently in a cloud development environment (Replit) while managing dynamic web ports.

🏆 Accomplishments that we're proud of

I am incredibly proud of building a fully functional, cryptographically secure cybersecurity tool using standard Python libraries! Creating a working SHA-256 chain that links every log entry together mathematically was a huge win for me.

🧠 What we learned

Through this project, I learned the core concepts of cybersecurity deception, network logging, and cryptographic data integrity. I learned how to analyze request headers to differentiate between a standard web browser and an automated attack script (like SQLmap or scanning bots).

🚀 What's next for Cyber Trap AI

In the future, I plan to add Geo-IP mapping so we can see the exact countries of attackers on a world map. I also want to build a feature that automatically blocks the attacker's IP on the local firewall once they trigger the honeypot traps!

Built With

• cryptography
• css3
• html5
• json
• python
• replit
• sqlite